39 (2011) 1154–1169

Contents lists available at ScienceDirect

Energy Policy

journal homepage: www.elsevier.com/locate/enpol

Providing all global energy with wind, water, and , Part I: Technologies, energy resources, quantities and areas of infrastructure, and materials

Mark Z. Jacobson a,n, Mark A. Delucchi b,1 a Department of Civil and Environmental Engineering, Stanford University, Stanford, CA 94305-4020, USA b Institute of Transportation Studies, University of at Davis, Davis, CA 95616, USA article info abstract

Article history: , pollution, and energy insecurity are among the greatest problems of our time. Addressing Received 3 September 2010 them requires major changes in our energy infrastructure. Here, we analyze the feasibility of providing Accepted 22 November 2010 worldwide energy for all purposes (, transportation, heating/cooling, etc.) from wind, Available online 30 December 2010 water, and sunlight (WWS). In Part I, we discuss WWS energy system characteristics, current and future Keywords: energy demand, availability of WWS resources, numbers of WWS devices, and area and material requirements. In Part II, we address variability, economics, and policy of WWS energy. We estimate that Solar power 3,800,000 5 MW wind turbines, 49,000 300 MW concentrated solar plants, 40,000 300 MW solar Water power PV power plants, 1.7 billion 3 kW rooftop PV systems, 5350 100 MW plants, 270 new 1300 MW hydroelectric power plants, 720,000 0.75 MW wave devices, and 490,000 1 MW tidal turbines can power a 2030 WWS world that uses electricity and electrolytic for all purposes. Such a WWS infrastructure reduces world power demand by 30% and requires only 0.41% and 0.59% more of the world’s land for footprint and spacing, respectively. We suggest producing all new energy with WWS by 2030 and replacing the pre-existing energy by 2050. Barriers to the plan are primarily social and political, not technological or economic. The energy cost in a WWS world should be similar to that today. & 2010 Elsevier Ltd. All rights reserved.

1. Introduction improving energy efficiency. Pacala and Socolow (2004) suggested a similar portfolio, but expanded it to include reductions in deforestation A solution to the problems of climate change, air pollution, water and conservation tillage and greater use of hydrogen in vehicles. pollution, and energy insecurity requires a large-scale conversion to More recently, Fthenakis et al. (2009) analyzed the technical, clean, perpetual, and reliable energy at low cost together with an geographical, and economic feasibility for solar energy to supply increase in energy efficiency. Over the past decade, a number of studies the energy needs of the U.S. and concluded (p. 397) that ‘‘it is clearly have proposed large-scale plans. Jacobson and feasible to replace the present fossil fuel energy infrastructure in

Masters (2001) suggested that the U.S. could satisfy its Kyoto Protocol the U.S. with solar power and other renewables, and reduce CO2 requirement for reducing carbon dioxide emissions by replacing 60% of emissions to a level commensurate with the most aggressive its generation with 214,000–236,000 wind turbines rated at climate-change goals’’. Jacobson (2009) evaluated several long- 1.5 MW (million watts). Also in 2001, Czisch (2006) suggested that a term energy systems according to environmental and other totally renewable electricity supply system, with intercontinental criteria, and found WWS systems to be superior to nuclear, transmission lines linking dispersed wind sites with hydropower fossil-fuel, and systems (see further discussion in Section 2). backup, could supply Europe, North Africa, and East Asia at total costs He proposed to address the hourly and seasonal variability of per kWh comparable with the costs of the current system. Hoffert et al. WWS power by interconnecting geographically disperse renew- (2002) suggested a portfolio of solutions for stabilizing atmospheric able energy sources to smooth out loads, using hydroelectric power

CO2, including increasing the use of renewable energy and nuclear to fill in gaps in supply. He also proposed using battery-electric energy, decarbonizing fossil fuels and sequestering carbon, and vehicles (BEVs) together with utility controls of electricity dispatch to them through smart meters, and storing electricity in hydrogen or solar-thermal storage media. Cleetus et al. (2009) subsequently n Corresponding author. Tel.: +1 650 723 6836. presented a ‘‘blueprint’’ for a clean-energy economy to reduce E-mail addresses: [email protected] (M.Z. Jacobson), [email protected] (M.A. Delucchi). CO2-equivalent GHG emissions in the U.S. by 56% compared 1 Tel.: +1 916 989 5566. with the 2005 levels. That study featured an economy-wide CO2

0301-4215/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.enpol.2010.11.040 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1155

Table 1 Recent studies of rapid, large-scale development of renewable energy.

Study by sector Time frame Geographic scope

This study and Jacobson and Delucchi (2009) Electricity transport heat/cool 100% WWS All new energy: 2030. World All energy: 2050 Alliance for Climate Protection (2009) Electricity transport 100% WWS+Bm 2020 U.S. Parsons-Brinckerhoff (2009) Electricity transport heat/cool 80% WWS+NCBmBf 2050 UK Price-Waterhouse-Coopers (2010) Electricity 100% WWS+Bm 2050 Europe & North Africa Beyond Zero Emissions (2010) Electricity transport heat/cool 100% WWS+Bm 2020 Australia European Climate Foundation (ECF) (2010) Electricity transport heat/cool 80% WWS+NCBm 2050 Europe European Renewable Energy Council (EREC) (April (2010) Electricity transport heat/cool 100% WWS+BmBf 2050 Europe

WWS¼wind, water, solar power; FF¼fossil fuels; Bm¼; Bf¼liquid ; N¼nuclear; C¼coal-CCS. Cleetus et al. (2009) is not included only because its focus is mainly on efficiency and demand management, with only modest increases in renewable energy. cap-and-trade program and policies to increase energy efficiency matches demand, the economics of WWS generation and transmis- and the use of renewable energy in industry, buildings, electricity, sion, the economics of the use of WWS power in transportation, and and transportation. Sovacool and Watts (2009) suggested that a policy issues. Although we recognize that a comprehensive plan to completely renewable electricity sector for New Zealand and the address global environmental problems must also address other is feasible. sectors, including agriculture (Horrigan et al., 2002; Wall and Smit, In Jacobson and Delucchi (2009), we outlined a large-scale plan 2005) and forestry (Niles et al., 2002), we do not address those to power the world for all purposes with WWS (no biofuels, nuclear issues here. power, or coal with carbon capture). The study found that it was technically feasible to power the world with WWS by 2030 but such a conversion would almost certainly take longer due to the 2. Clean, low-risk, sustainable energy systems difficulty in implementing all necessary policies by then. However, we suggested, and this study reinforces, the concept that all new 2.1. Evaluation of long-term energy systems: why we choose WWS energy could be supplied by WWS by 2030 and all existing energy power could be converted to WWS by 2050. The analysis presented here is an extension of that work. Because climate change (particularly loss of the Arctic sea ice Table 1 compares and summarizes several other recent large- cap), air pollution, and energy insecurity are the current and scale plans. While all plans are ambitious, forward thinking, and growing problems, but it takes several decades for new technol- detailed, they differ from our plan, in that they are for limited world ogies to become fully adopted, we consider only options that have regions and none relies completely on WWS. However, some come been demonstrated in at least pilot projects and that can be scaled close in the electric power sector, relying on only small amounts of up as part of a global energy system without further major non-WWS energy in the form of biomass for electric power technology development. We avoid options that require substan- production. Those studies, however, address only electricity and/ tial further technological development and that will not be ready to or transport, but not heating/cooling. begin the scale-up process for several decades. Note that we select More well known to the public than the scientific studies, technologies based on the state of development of the technology perhaps, are the ‘‘Repower America’’ plan of former Vice-President only rather than whether industrial capacity is currently ramped and Nobel-Peace Prize winner , and a similar proposal by up to produce the technologies on a massive scale or whether businessman T. Boone Pickens. Mr. Gore’s proposal calls for society is motivated to change to the technologies. In this paper and improvements in energy efficiency, expansion of renewable in Part II, we do consider the feasibility of implementing the chosen energy generation, modernization of the transmission grid, and the technologies based on estimated costs, necessary policies, and conversion of motor vehicles to electric power. The ultimate (and available materials as well as other factors. ambitious) goal is to provide America ‘‘with 100% clean electricity In order to ensure that our energy system remains clean even within 10 years,’’ which Mr. Gore proposes to achieve by increasing with large increases in population and economic activity in the long the use of wind and concentrated solar and improving energy run, we consider only those technologies that have essentially zero efficiency (Alliance for Climate Protection, 2009). In Gore’s plan, emissions of greenhouse gases and air pollutants per unit of output solar PV, geothermal, and biomass electricity would grow only over the whole ‘‘lifecycle’’ of the system. Similarly, we consider only modestly, and and would not grow. those technologies that have low impacts on wildlife, water Mr. Pickens’ plan is to obtain up to 22% of the U.S. electricity from pollution, and land, do not have significant waste-disposal or wind, add solar capacity to that, improve the electric grid, increase terrorism risks associated with them, and are based on primary energy efficiency, and use instead of oil as a transitional resources that are indefinitely renewable or recyclable. fuel (Pickens, 2009). The previous work by Jacobson (2009) indicates that WWS There is little doubt that the large-scale use of renewable energy power satisfies all of these criteria. He ranked several long-term envisaged in these plans and studies would greatly mitigate or energy systems with respect to their impacts on global warming, eliminate a wide range of environmental and human health air pollution, water supply, land use, wildlife, thermal pollution, impacts of energy use (e.g., Jacobson, 2009; Sovacool and water–chemical pollution, and nuclear weapons proliferation. The Sovacool, 2009; Colby et al., 2009; Weisser, 2007; Fthenakis and ranking of electricity options, starting with the highest, included: Kim, 2007). But, is a large-scale transformation of the world’s wind power, concentrated solar, geothermal, tidal, solar photo- energy systems feasible? In this paper and in Part II, we address this voltaic, wave, and hydroelectric power, all of which are powered by question by examining the characteristics and benefits of wind, wind, water, or sunlight (WWS). He also found that the use of BEVs water, and solar (WWS)-energy systems, the availability of WWS and hydrogen fuel-cell vehicles (HFCVs) powered by the WWS resources, supplies of critical materials, methods of addressing the options would largely eliminate pollution from the transportation variability of WWS energy to ensure that power supply reliably sector. Here, we consider these technologies and other existing 1156 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 technologies for the heating/cooling sectors, discussed in Section 2. installed, that would double the current installations of nuclear Although other clean WWS electric power sources, such as ocean or power worldwide. Many more countries would possess nuclear river current power, could be deployed in the short term, these are facilities, increasing the likelihood that these countries would use not examined here simply because we could not cover every the facilities to hide the development of nuclear weapons as has technology. Nevertheless, we do cover related although slightly occurred historically. different power sources (e.g., wave, tidal, and hydroelectric power). Second, nuclear energy results in 9–25 times more carbon Finally, Jacobson (2009) concluded that coal with carbon emissions than wind energy, in part due to emissions from uranium capture, corn ethanol, cellulosic ethanol, and nuclear power were refining and transport and reactor construction (e.g., Lenzen, 2008; all moderately or significantly worse than WWS options with Sovacool, 2008), in part due to the longer time required to site, respect to environmental and land use impacts. Similarly, here we permit, and construct a nuclear plant compared with a do not consider any combustion sources, such as coal with carbon (resulting in greater emissions from the fossil-fuel electricity sector capture, corn ethanol, cellulosic ethanol, soy , algae during this period; Jacobson, 2009), and in part due to the greater biodiesel, biomass for electricity, other biofuels, or natural gas, loss of soil carbon due to the greater loss in vegetation resulting because none of these technologies can reduce GHG and from covering the ground with nuclear facilities relative to wind air-pollutant emissions to near zero, and all can have significant turbine towers, which cover little ground. Although recent con- problems in terms of land use, water use, or resource availability struction times worldwide are shorter than the 9-year median (See Delucchi (2010) for a review of land-use, climate-change, and construction times in the U.S. since 1970 (Koomey and Hultman, water-use impacts of biofuels.) For example, even the most 2007), they still averaged 6.5 years worldwide in 2007 (Ramana, climate-friendly and ecologically acceptable sources of ethanol, 2009), and this time must be added to the site permit time such as unmanaged, mixed grasses restored to their native (non- (3 years in the U.S.) and construction permit and issue time agricultural) habitat (Tilman et al., 2006), will cause air pollution (3 years). The overall historic and present range of nuclear plan- mortality on the same order as gasoline (Jacobson, 2007; Anderson, ning-to-operation times for new nuclear plants has been 11–19 years, 2009; Ginnebaugh et al., 2010). The use of carbon capture and compared with an average of 2–5 years for wind and solar installations sequestration (CCS) can reduce CO2 emissions from the stacks of (Jacobson, 2009). Feiveson (2009) observes that ‘‘because wind tur- coal power plants by 85–90% or more, but it has no effect on CO2 bines can be installed much faster than could nuclear, the cumulative emissions due to the mining and transport of coal; in fact it will greenhouse gas savings per capital invested appear likely to be greater increase such emissions and of air pollutants per unit of net for wind’’ (p. 67). The long time required between planning and delivered power and will increase all ecological, land-use, operation of a nuclear power plant poses a significant risk to the Arctic air-pollution, and water-pollution impacts from coal mining, sea ice. Sea ice records indicate a 32% loss in the August 2010 sea ice transport, and processing, because the CCS system requires 25% area relative to the 1979–2008 mean (Cryosphere Today, 2010). Such more energy, thus 25% more coal combustion, than does a system rapid loss indicates that solutions to global warming must be without CCS (IPCC, 2005). implemented quickly. Technologies with long lead times will allow For several reasons we do not consider nuclear energy the high-albedo Arctic ice to disappear, triggering more rapid positive (conventional fission, breeder reactors, or fusion) as a long-term feedbacks to warmer temperatures by uncovering the low-albedo global energy source. First, the growth of nuclear energy has ocean below. historically increased the ability of nations to obtain or enrich Third, conventional nuclear fission relies on finite stores of uranium for nuclear weapons (Ullom, 1994), and a large-scale uranium that a large-scale nuclear program with a ‘‘once through’’ worldwide increase in nuclear energy facilities would exacerbate fuel cycle would exhaust in roughly a century (e.g., Macfarlane and this problem, putting the world at greater risk of a nuclear war or Miller, 2007; Adamantiades and Kessides, 2009). In addition, terrorism catastrophe (Kessides, 2010; Feiveson, 2009; Miller and accidents at nuclear power plants have been either catastrophic Sagan, 2009; Macfarlane and Miller, 2007; Harding, 2007). The (Chernobyl) or damaging (Three-Mile Island), and although historic link between energy facilities and weapons is evidenced by the nuclear industry has improved the safety and performance the development or attempted development of weapons capabil- of reactors, and has proposed new (but generally untested) ities secretly in nuclear energy facilities in Pakistan, India ‘‘inherently’’ safe reactor designs (Piera, 2010; Penner et al., (Federation of American Scientists, 2010), Iraq (prior to 1981), Iran 2008; Adamantiades and Kessides, 2009; Mourogov et al., 2002; (e.g., Adamantiades and Kessides, 2009, p. 16), and to some extent Mourogov, 2000), there is no guarantee that the reactors will be North Korea. Feiveson (2009) writes that ‘‘it is well understood that designed, built, and operated correctly. For example, Pacific Gas one of the factors leading several countries now without nuclear and Electric Company had to redo some modifications it made to its power programs to express interest in nuclear power is the Diablo Canyon nuclear power plant after the original work was foundation that such programs could give them to develop done backwards (Energy Net, 2010), and French nuclear regulators weapons’’ (p. 65). Kessides (2010) asserts, ‘‘a robust global expan- recently told the firm Areva to correct a safety design flaw in its sion of civilian nuclear power will significantly increase prolifera- latest-generation reactor (Nuclear Power Daily, 2009). Further, tion risks unless the current non-proliferation regime is catastrophic scenarios involving terrorist attacks are still concei- substantially strengthened by technical and institutional measures vable (Feiveson, 2009). Even if the risks of catastrophe are very and its international safeguards system adequately meets the new small, they are not zero (Feiveson, 2009), whereas with wind and challenges associated with a geographic spread and an increase in solar power, the risk of catastrophe is zero. Finally, conventional the number of nuclear facilities’’ (p. 3860). Similarly, Miller and nuclear power produces , which must be stored Sagan (2009) write, ‘‘it seems almost certain that some new for thousands of years, raising technical and long-term cost entrants to nuclear power will emerge in the coming decades questions (Barre´, 1999; von Hippel, 2008; Adamantiades and and that the organizational and political challenges to ensure the Kessides, 2009). safe and secure spread of nuclear technology into the developing ‘‘Breeder’’ nuclear reactors have similar problems as conven- world will be substantial and potentially grave’’ (p. 12). tional fission reactors, except that they produce less low-level If the world were converted to electricity and electrolytic radioactive waste than do conventional reactors and re-use the hydrogen by 2030, the 11.5 TW in resulting power demand would spent fuel, thereby extending uranium reserves, perhaps indefi- require 15,800 850 MW nuclear power plants, or one installed nitely (Penner et al., 2008; Purushotham et al., 2000; Till et al., every day for the next 43 years. Even if only 5% of these were 1997). However, they produce nuclear material closer to weapons M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1157 grade that can be reprocessed more readily into nuclear weapons 2.2. Characteristics of electricity-generating WWS technologies (Kessides, 2010; Adamantiades and Kessides, 2009; Macfarlane and Miller, 2007; Glaser and Ramana, 2007), although some 2.2.1. Wind technologies have technical features that make diversion and Wind turbines convert the energy of the wind into electricity. reprocessing especially difficult—albeit not impossible (Hannum Generally, a gearbox turns the slow-moving turbine rotor into et al., 1997; Kessides, 2010; Penner et al., 2008). Kessides (2010) faster-rotating gears, which convert mechanical energy to elec- writes, ‘‘analyses of various reactor cycles have shown that all have tricity in a generator. Some modern turbines are gearless. Although some potential for diversion, i.e., there is no proliferation-proof less efficient, small turbines can be used in homes or buildings. nuclear power cycle’’ (p. 3861). Wind farms today appear on land and offshore, with individual A related proposal is to use thorium as a nuclear fuel, which is turbines ranging in size up to 7 MW, with 10 MW planned. less likely to lead to nuclear weapons proliferation than the use of High-altitude wind energy capture is also being pursued today uranium, produces less long-lived radioactive waste, and greatly by several companies. extends uranium resources (Macfarlane and Miller, 2007). However, thorium reactors require the same significant time lag between planning and operation as conventional uranium reactors 2.2.2. Wave and most likely longer because few developers and scientists have Winds passing over water create surface waves. The faster the experience with constructing or running thorium reactors. As such, wind speed, the longer the wind is sustained, the greater the this technology will result in greater emissions from the back- distance the wind travels, the greater the wave height, and the ground electric grid compared with WWS technologies, which have greater the wave energy produced. devices capture a shorter time lag. In addition, lifecycle emissions of carbon from a energy from ocean surface waves to produce electricity. One type of thorium reactor are on the same order as those from a uranium device is a buoy that rises and falls with a wave. Another type is a reactor. Further, thorium still produces radioactive waste contain- surface-following device, whose up-and-down motion increases ing 231Pa, which has a half-life of 32,760 years. It also produces 233U, the pressure on oil to drive a hydraulic motor. which can be used in fission weapons, such as in one nuclear bomb core during the Operation Teapot nuclear tests in 1955. Weapo- 2.2.3. Geothermal nization, though, is made more difficult by the presence of 232U. Steam and hot water from below the Earth’s surface have been Fusion of light atomic nuclei (e.g., protium, deuterium, or tritium) used historically to provide heat for buildings, industrial processes, theoretically could supply power indefinitely without long-lived and domestic water and to generate electricity in geothermal radioactive wastes as the products are isotopes of helium (Ongena power plants. In power plants, two boreholes are drilled—one for and Van Oost, 2006; Tokimatsu et al., 2003); however, it would steam alone or liquid water plus steam to flow up, and the second produce short-lived waste that needs to be removed from the reactor for condensed water to return after it passes through the plant. In core to avoid interference with operations, and it is unlikely to be some plants, steam drives a turbine; in others, hot water heats commercially available for at least another 50–100 years (Tokimatsu another fluid that evaporates and drives the turbine. et al., 2003; Barre´, 1999; Hammond, 1996), long after we will have needed to transition to alternative energy sources. By contrast, wind and solar power are available today, will last indefinitely, and pose 2.2.4. Hydroelectricity no serious risks. Note that our reasons for excluding nuclear are not Water generates electricity when it drops gravitationally, driv- economic. A brief discussion of the economics of nuclear power is ing a turbine and generator. While most hydroelectricity is given in Appendix A. produced by water falling from dams, some is produced by water For these reasons, we focus on WWS technologies. We assume flowing down rivers (run-of-the-river electricity). that WWS will supply electric power for the transportation, heating (including high-temperature heating and cooking)/cooling sectors, which traditionally have relied mainly on the direct use of oil or gas 2.2.5. Tidal rather than electricity, as well as for traditional electricity-con- A tidal turbine is similar to a in that it consists of a suming end uses such as lighting, cooling, manufacturing, motors, rotor that turns due to its interaction with water during the ebb and electronics, and telecommunications. Although we focus mainly on flow of a tide. Tidal turbines are generally mounted on the sea floor. , we acknowledge and indeed emphasize the impor- Since tides run about 6 h in one direction before switching tance of demand-side energy conservation measures to reduce the directions for 6 h, tidal turbines can provide a predictable energy requirements and impacts of energy supply. Demand-side energy- source. O’Rourke et al. (2010) provide an excellent overview of the conservation measures include improving the energy-out/energy- technology of tidal energy. in efficiency of end uses (e.g., with more efficient vehicles, more efficient lighting, better insulation in homes, and the use of heat- 2.2.6. Solar PV exchange and filtration systems), directing demand to low-energy- Solar photovoltaics (PVs) are arrays of cells containing a use modes (e.g., using public transit or telecommuting instead of material, such as silicon, that converts solar radiation into elec- driving), large-scale planning to reduce energy demand without tricity. Today, solar PVs are used in a wide range of applications, compromising economic activity or comfort (e.g., designing cities from residential rooftop power generation to medium-scale utility- to facilitate greater use of non-motorized transport and to have level power generation. better matching of origins and destinations, thereby reducing the need for travel), and designing buildings to use solar energy directly (e.g., with more daylighting, solar hot water heating, and improved 2.2.7. CSP passive solar heating in winter and cooling in summer). For a Concentrated solar power (CSP) systems use mirrors or reflec- general discussion of the potential to reduce energy use in tive lenses to focus sunlight on a fluid to heat it to a high transportation and buildings, see the American Physical Society temperature. The heated fluid flows from the collector to a heat (2008). For a classification scheme that facilitates analyses of the engine where a portion of the heat is converted to electricity. Some potential gains from energy efficiency, see Cullen and Allwood types of CSP allow the heat to be stored for many hours so that (2009). electricity can be produced at night. 1158 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169

2.3. Use of WWS power for transportation 2.4. Use of WWS power for heating and cooling

Transportation technologies that must be deployed on a large For building water and air heating using WWS power, we scale to use WWS-power include primarily battery-electric vehi- propose the use of air- and ground-source heat-pump water and air cles (BEVs), hydrogen fuel-cell vehicles (HFCVs), and hybrid heaters and electric resistance water and air heaters. Heat pump air BEV-HFCVs. For ships, we propose the use of hybrid hydrogen fuel heaters also can be used in reverse for air conditioning. These cell-battery systems, and for aircraft, liquefied hydrogen combus- technologies exist today although in most places they satisfy less tion (Appendix A). demand than do natural gas or oil-fired heaters. The use of BEVs store electricity in and draw power from batteries to run an electricity for heating and cooking, like the use of electricity for that drives the vehicle. So long as the electricity source transportation, is most beneficial when the electricity comes from is clean, the BEV system will have zero emissions of air pollutants and WWS. For high-temperature industrial processes, we propose that greenhouse gases over the entire energy lifecycle—something that energy be obtained by combustion of electrolytic hydrogen internal-combustion-engine vehicles (ICEVs) using liquid fuels cannot (Appendix A). achieve. Moreover, BEVs provide up to 5 times more work in distance traveled per unit of input energy than do ICEVs (km/kWh-outlet versus km/kWh-gasoline). BEVs have existed for decades in small 3. Energy resources needed and available levels of production, but today most major automobile companies are developing BEVs. The latest generation of vehicles uses lithium-ion The power required today to satisfy all end uses worldwide is batteries, which do not use the toxic chemicals associated with lead- about 12.5 trillion watts (TW) (EIA, 2008a; end-use energy only, acid or the nickel-cadmium batteries. excludes losses in production and transmission). In terms of Hydrogen vehicles (HFCVs) use a fuel cell to convert primary energy, about 35% is from oil, 27% from coal, 23% from hydrogen fuel and oxygen from the air into electricity that is used to natural gas, 6% from nuclear, and the rest from biomass, sunlight, run an electric motor. HFCVs are truly clean only if the hydrogen is wind, and geothermal. Delivered electricity is a little over 2 TW of produced by passing WWS-derived electricity through water the end-use total. (electrolysis). Thus, we propose producing hydrogen only in this The EIA (2008a) projects that in the year 2030, the world will way. Several companies have prototype HFCVs, and California had require almost 17 TW in end-use power, and the U.S. almost 3 TW about 200 HFCVs on the road in 2009 (California Fuel Cell (Table 2). They also project that the breakdown in terms of primary Partnership, 2009). Hydrogen fueling stations, though, are practi- energy in 2030 will be similar to that today—heavily dependent on cally non-existent and most hydrogen today is produced by steam- fossil fuels, and hence almost certainly unsustainable. What would reforming of natural gas, which is not so clean as hydrogen world power demand look like if instead a sustainable WWS produced by WWS-electrolysis. system supplied all end-use energy needs?

Table 2 Projected end-use power in 2030, by sector, U.S. and the world, conventional fossil-fuel case, and replacing 100% of fossil fuel and wood combustion with WWS.

Energy sector, by EIA TW power in 2030 (conventional Elect. fract. End-use energy/work w.r.t. Upstream EHCM TW power in 2030 replacing energy-use categories fossil fuels) fossil fuel factor factor all fossil fuels with WWS

World U.S. Electric e-H2 World U.S.

Residential Liquids 0.37 0.04 0.95 0.82 1.43 1.00 0.90 0.29 0.03 Natural gas 0.84 0.18 0.95 0.82 1.43 1.00 0.90 0.61 0.13 Coal 0.11 0.00 1.00 0.82 1.43 1.00 0.90 0.08 - Electricity 0.92 0.20 1.00 1.00 1.00 1.00 0.95 0.83 0.18 Renewables 0.02 0.01 0.50 0.82 1.43 1.00 0.90 0.02 0.01 Total 2.26 0.43 1.83 0.35

Commercial Liquids 0.18 0.02 0.90 0.82 1.43 1.00 0.95 0.15 0.02 Natural gas 0.32 0.13 0.90 0.82 1.43 1.00 0.95 0.26 0.10 Coal 0.03 0.00 0.90 0.82 1.43 1.00 0.95 0.03 0.00 Electricity 0.78 0.22 1.00 1.00 1.00 1.00 1.00 0.78 0.22 Renewables 0.01 0.00 0.90 0.82 1.43 1.00 0.95 0.01 0.00 Total 1.32 0.38 1.22 0.35

Industrial Liquids 2.41 0.31 0.60 0.82 1.43 0.72 0.95 1.76 0.22 Natural gas 2.35 0.28 0.60 0.82 1.43 0.82 0.95 1.95 0.23 Coal 2.15 0.08 0.60 0.82 1.43 0.73 0.95 1.59 0.06 Electricity 1.75 0.12 1.00 1.00 1.00 0.93 1.00 1.62 0.11 Renewables 0.15 0.14 0.90 0.82 1.43 1.00 0.95 0.13 0.12 Total 8.80 0.92 7.05 0.74

Transportation Liquids 4.44 1.07 0.73 0.19 0.64 1.18 0.85 1.30 0.31 Natural gas 0.05 0.03 0.90 0.82 1.43 1.00 0.85 0.04 0.02 Coal – 0.00 0.90 0.82 1.43 1.00 0.85 – – Electricity 0.04 0.00 1.00 1.00 1.00 1.00 0.95 0.03 – Total 4.53 1.10 1.37 0.33 Total end uses 16.92 2.83 11.47 1.78

Notes: see Appendix A.2. M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1159

Table 2 shows our estimates of global and U.S. end-use energy plus ocean surfaces at 100 m if all wind at all speeds were used to demand, by sector, in a world powered entirely by WWS, with zero power wind turbines (Table 3); however, the wind power over land fossil-fuel and biomass combustion. We have assumed that all end in locations over land and near shore where the wind speed is 7 m/s uses that feasibly can be electrified use WWS power directly, and or faster (the speed necessary for cost-competitive wind energy) is that the remaining end uses use WWS power indirectly in the form around 72–170 TW (Archer and Jacobson, 2005; Lu et al., 2009; of electrolytic hydrogen (hydrogen produced by splitting water Fig. 1). Over half of this power is in locations that could practically with WWS power). As explained in Section 2 we assume that most be developed. Large regions of fast winds worldwide include the uses of fossil fuels for heating/cooling can be replaced by electric of the U.S. and Canada, Northern Europe, the Gobi heat pumps, and that most uses of liquid fuels for transportation and Sahara Deserts, much of the Australian desert areas, and can be replaced by BEVs. The remaining, non-electric uses can be parts of South Africa and Southern South America and South Africa. supplied by hydrogen, which we assume would be compressed for In the U.S., wind from the Great Plains and offshore the East use in fuel cells in remaining non-aviation transportation, liquefied Coast (Kempton et al., 2007) could supply all U.S. energy needs. and combusted in aviation, and combusted to provide heat directly Other windy offshore regions include the North Sea, the West Coast in the industrial sector. The hydrogen would be produced using of the U.S. (Dvorak et al., 2010), and the East Coast of Asia among WWS power to split water; thus, directly or indirectly, WWS others. powers the world. Extraction from the wind of 100% of the power needed for the As shown in Table 2, the direct use of electricity, for example, for world in 2030 (11.5 TW from Table 2) would reduce the overall heating or electric motors, is considerably more efficient than is power in the wind at 100 m by o1% (Santa Maria and Jacobson, fuel combustion in the same application. The use of electrolytic 2009). Such extracted power is eventually dissipated to heat, a hydrogen is less efficient than is the use of fossil fuels for direct portion of which is cycled back to produce more potential energy, heating but more efficient for transportation when fuel cells are which produces kinetic energy, regenerating some of the wind. used; the efficiency difference between direct use of electricity and The remaining heat goes toward slightly increasing air and ground electrolytic hydrogen is due to the energy losses of electrolysis, and temperature, but this addition is very small. For example, the in the case of most transportation uses, the energy requirements of maximum additional radiative forcing due to powering the world compression and the greater inefficiencies of fuel cells than with wind is 11.5 TW/5.106 1014 m2 (area of the Earth)¼0.022 batteries. Assuming that some additional modest energy-conser- W/m2, which is only 0.7% of the 3W/m2 forcing due to all vation measures are implemented (see the list of demand-side conservation measures in Section 2) and subtracting the energy Annual wind speed 100m above topo (m/s) requirements of refining, we estimate that an all-WWS (global: 7.0; l and: 6.1; sea: 7.3) world would require 30% less end-use power than the EIA 90 projects for the conventional fossil-fuel scenario (Table 1). 10 How do the energy requirements of a WWS world, shown in Table 2, compare with the availability of WWS power? Table 3 gives 8 the estimated power available worldwide from renewable energy, in terms of raw resources, resources available in high-energy 0 6 locations, resources that can feasibly be extracted in the near term considering cost and location, and the current resources used. The table indicates that only solar and wind can provide more power on 4 their own than energy demand worldwide. Wind in developable locations can power the world about 3–5 times over and solar, -90 2 about 15–20 times over. -180 -90 0 90 180 Fig. 1 shows the modeled world wind resources at 100 m, in the Fig. 1. Map of the yearly averaged world wind speed (m/s) at 100 m above sea level range of the hub height of modern wind turbines. Globally, at 1.5 1.51 resolution, generated with the GATOR-GCMOM 3-D global model 1700 TW of wind energy are available over the world’s land (Jacobson, 2010).

Table 3 Power available in energy resource worldwide if the energy is used in conversion devices, in locations where the energy resource is high, in likely-developable locations, and in delivered electricity in 2005 or 2007 (for wind and solar PV).

Energy technology Power worldwide (TW) Power in high-energy Power in likely- Current power delivered locations (TW) developable locations (TW) as electricity (TW)

Wind 1700a 72–170b 40–85c 0.02d Wave 42.7d 2.7e 0.5d 0.000002d Geothermal 45f 2g 0.07-0.14d 0.0065d Hydroelectric 1.9d o1.9d 1.6d 0.32d Tidal 3.7d 0.8d 0.02d 0.00006d Solar PV 6500h 1300i 340d 0.0013d CSP 4600j 920j 240j 000046d

a Fig. 1 here; accounts for all wind speeds at 100 m over land and ocean. b Locations over land or near the coast where the mean wind speedZ7 m/s at 80 m (Archer and Jacobson, 2005) and at 100 m (Lu et al, 2009; Fig. 1 here). c Eliminating remote locations. d Jacobson (2009) and references therein. e Wave power in coastal areas. f Fridleifsson et al. (2008). g Includes estimates of undiscovered reservoirs over land. h Fig. 2 here, assuming use of 160 W solar panels and areas determined in Jacobson (2009), over all latitudes, land, and ocean. i Same as (h) but locations over land between 50S and 50N. j Scaling solar PV resource with relative land area requirements from Jacobson (2009). 1160 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 greenhouse gases. Since wind turbines replace other electricity suggesting air-cooled plants may be a viable alternative in water- sources that also produce heat in this manner (Santa Maria and limited locations. Jacobson, 2009), wind turbines (and other renewable electricity The other WWS technologies have much less resource avail- sources) replacing current infrastructure cause no net heat addition ability than do wind, CSP, and PV (Table 3), yet can still contribute to the atmosphere. They serve only to reduce global-warming beneficially to the WWS solution. Wave power can be extracted pollutants and heath-affecting air pollutants that current electri- practically only near coastal areas, which limits its worldwide city and energy sources produce. potential. Although the Earth has a very large reservoir of geother- Fig. 2 shows the distribution of solar energy at the Earth’s mal energy below the surface, most of it is too deep to extract surface. Globally, 6500 TW of solar energy are available over the practically. Even though hydroelectric power today exceeds all world’s land plus ocean surfaces if all sunlight is used to power other sources of WWS power, its future potential is limited because photovoltaics (Table 3); however, the deliverable solar power over most of the large reservoirs suitable for generating hydropower are land in locations where solar PV could practically be developed is already in use. about 340 TW. Alternatively CSP could provide about 240 TW of the Further, although there is enough feasibly developable wind world’s power output, less than PV since the land area required for and solar power to supply the world, other WWS resources will be CSP without storage is about one-third greater than is that for PV. more abundant and more economical than wind and solar in many With thermal storage, the land area for CSP increases since more locations. Finally, wind and solar power are variable, so geothermal solar collectors are needed to provide energy for storage, but and , which provide relatively constant power, and energy output does not change and the energy can be used at night. hydroelectric, which fills in gaps, will be important for providing a However, water-cooled CSP plants can require water for cooling stable electric power supply. during operation (about 8 gal/kWh—much more than PVs and See a detailed discussion of this in Part II of this work, Delucchi wind (0 gal/kWh), but less than nuclear and coal (40 gal/kWh) and Jacobson (this issue). (Sovacool and Sovacool, 2009)), and this might be a constraint in some areas. This constraint is not accounted for in the estimates of Table 3. However, air-cooled CSP plants require over 90% less 4. Quantities and areas of plants and devices required water than water-cooled plants at the cost of only about 5% less electric power and 2–9% higher electricity rates (USDOE, 2008b), How many WWS power plants or devices are required to power the world and U.S.? Table 4 provides an estimate for 2030, assuming a given fractionation of the demand (from Table 2) among technologies. Wind and solar together are assumed to comprise 90% of the future supply based on their relative abun- Surface downward solar radiation (W/m2) dances (Table 3). Although 4% of the proposed future supply is (global avg: 193; land: 183) 90 hydro, most of this amount (70%) is already in place. Solar PV is 250 divided into 30% rooftop, based on an analysis of likely available rooftop area (Jacobson, 2009) and 70% power plant. Rooftop PV has three major advantages over power-plant PV: rooftop PV does not 200 require an electricity transmission and distribution network, it can 0 be integrated into a hybrid solar system that produces heat, light, and electricity for use on site (Chow, 2010), and it does not require 150 new land area. Table 4 suggests that almost 4 million 5 MW wind turbines (over land or water) and about 90,000 300 MW PV plus 100 CSP power plants are needed to help power the world. Already, -90 about 0.8% of the wind is installed. -180 -90 0 90 180 The total footprint on the ground (for the turbine tubular tower and base) for the 4 million wind turbines required to power 50% of Fig. 2. Map of the yearly averaged downward surface solar radiation reaching the 2 surface (W/m2) at 1.5 1.51 resolution, generated with the GATOR-GCMOM 3-D the world’s energy is only 48 km , smaller than Manhattan 2 global model (Jacobson, 2010). (59.5 km ) whereas the spacing needed between turbines to

Table 4 Number of WWS power plants or devices needed to power the world and U.S. total energy demand in 2030 (11.5 and 1.8 TW, respectively, from Table 2), assuming a given partitioning of the demand among plants or devices. Also shown are the footprint and spacing areas required to power the world, as a percentage of the global land area, 1.446 108 km2. Derived from appendix A of Jacobson (2009).

Energy technology Rated power of Percent of 2030 power Number of plants or Footprint area (% of Spacing area (% of Number of plants or one plant or demand met by plant/ devices needed global land area) global land area) devices needed U.S. device (MW) device World

Wind turbine 5 50 3.8 million 0.000033 1.17 590,000 Wave device 0.75 1 720,000 0.00026 0.013 110,000 Geothermal plant 100 4 5350 0.0013 0 830 Hydroelectric plant 1300 4 900a 0.407a 0 140a Tidal turbine 1 1 490,000 0.000098 0.0013 7600 Roof PV system 0.003 6 1.7 billion 0.042b 0 265 million Solar PV plant 300 14 40,000 0.097 0 6200 CSP plant 300 20 49,000 0.192 0 7600 Total 100 0.74 1.18 Total new land 0.41c 0.59c

a About 70% of the hydroelectric plants are already in place. See Jacobson (2009) for a discussion of apportioning the hydroelectric footprint area by use of the reservoir. b The footprint area for rooftop solar PV does not represent an increase in land since the rooftops already exist and are not used for other purposes. c Assumes 50% of the wind is over water, wave and tidal are in water, 70% of hydroelectric is already in place, and rooftop solar does not require new land. M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1161 minimize the effects of one turbine reducing energy to other PVs, CSP systems, BEVs, and electrolytic-HFCVs. In this section, we turbines is 1.17% of the global land area. The spacing can be used examine whether any of these technologies use materials that for agriculture, rangeland, open space, or can be open water. either are scarce or else concentrated in a few countries and hence Whereas, wind turbines have foundations under the ground larger subject to price and supply manipulation. than their base on the ground, such underground foundation areas are not footprint, which is defined as the area of a device or plant touching the top surface of the soil, since such foundations are 5.1. Wind power covered with dirt, allowing vegetation to grow and wildlife to flourish on top of them. The footprint area for wind also does not The primary materials needed for wind turbines include steel need to include temporary or unpaved dirt access roads, as most (for towers, nacelles, rotors, etc.), pre-stressed concrete (for large-scale wind will go over areas such as the Great Plains and towers), magnetic materials (for gearboxes), aluminum (nacelles), some desert regions, where photographs of several farms indicate copper (nacelles), wood epoxy (rotor blades), glassfiber reinforced unpaved access roads blend into the natural environment and are plastic (GRP) (for rotor blades), and carbon-filament reinforced often overgrown by vegetation. Offshore wind does not require plastic (CFRP) (for rotor blades). In the future, use of composites of roads at all. In farmland locations, most access roads have dual GFRP, CFRP, and steel will likely increase. purposes, serving agricultural fields as well as turbines. In cases The manufacture of four million 5 MW or larger wind turbines where paved access roads are needed, 1 km2 of land provides will require large amounts of bulk materials such as steel and 200 km (124 miles) of linear roadway 5 m wide, so access roads concrete (USDOE, 2008a). However, there do not appear to be would not increase the footprint requirements of wind farms more significant environmental or economic constraints on expanded than a small amount. The footprint area also does not include production of these bulk materials. The major components of transmission, since the actual footprint area of a transmission concrete – gravel, sand, and limestone – are widely abundant, and tower is smaller than the footprint area of a wind turbine. This is concrete can be recycled and re-used. The Earth does have a because a consists of four narrow metal somewhat limited reserves of economically recoverable iron ore support rods separated by distance, penetrating the soil to an (on the order of 100–200 years at current production rates (USGS, underground foundation. Many photographs of transmission 2009, p. 81)), but the steel used to make towers, nacelles, and rotors towers indicate more vegetation growing under the towers than for wind turbines should be virtually 100% recyclable (for example, around the towers since areas around that towers are often in the U.S. in 2007, 98% of steel construction beams and plates were agricultural or otherwise used land whereas the area under the recycled (USGS, 2009, p. 84)). The USDOE (2008a) concludes that tower is vegetated soil. Since the land under transmission towers the development of 20% wind energy by 2030 is not likely to be supports vegetation and wildlife, it is not considered footprint constrained by the availability of bulk materials for wind turbines. beyond the small area of the support axess roads. For wind power, the most problematic materials may be rare For non-rooftop solar PV plus CSP, the areas required are earth elements (REEs) like neodymium (Nd) used in permanent considered here to be entirely footprint although technically a magnets (PMs) in generators (Margonelli, 2009; Gorman, 2009; walking space, included here as footprint, is required between solar Lifton, 2009). In some wind-power development scenarios, panels (Jacobson, 2009). Powering 34% of the world with non- demand for REEs might strain supplies or lead to dependence on rooftop solar PV plus CSP requires about one-quarter of the land potentially insecure supplies. (e.g., Margonelli, 2009; Hurst, 2010). area for footprint plus spacing as does powering 50% of the world One estimate suggests that current PM generators in large wind with wind but a much larger footprint area alone than does wind turbines use 0.2 kg Nd/kWh, or one-third the 0.6 kg/kWh of an (Table 4). The footprint area required for rooftop solar PV has Nd-based permanent magnet (Hatch, 2009). Building the 19 already been developed, as rooftops already exist. As such, these million installed MW of wind power needed to power 50% of areas do not require further increases in land requirements. world energy in 2030 (Table 4) would require 3.8 million metric Geothermal power requires a smaller footprint than does solar tonnes of Nd, or about 4.4 million metric tonnes of Nd oxide (based but a larger footprint than does wind per unit energy generated. on Nd2O3), which would amount to approximately 100,000 metric The footprint area required for hydroelectric is large due to the tons of Nd oxide per year over a 40–50 year period. In 2008, the large area required to store water in a reservoir, but 70% of the world produced 124,000 metric tonnes of rare-earth oxide equiva- needed hydroelectric power for a WWS system is already in place. lent, which included about 22,000 metric tonnes of Nd oxide Together, the entire WWS solution would require the equivalent (Table 5). Annual world production of Nd therefore would have to of 0.74% of the global land surface area for footprint and 1.18% for increase by a factor of more than five to accommodate the demand spacing (or 1.9% for footprint plus spacing). Up to 61% of the for Nd for production of PMs for wind-turbine generators for our footprint plus spacing area could be over the ocean if all wind were global WWS scenario. placed over the ocean although a more likely scenario is that The global Nd reserve or resource base could support 122,000 30–60% of wind may ultimately be placed over the ocean given the metric tonnes of Nd oxide production per year (the amount needed strong wind speeds there (Fig. 1). If 50% of wind energy were over for wind generators in our scenario, plus the amount needed to the ocean, and since wave and tidal are over the ocean, and if we supply other demand in 2008) for at least 100 years, and perhaps consider that 70% of hydroelectric power is already in place and for several hundred years, depending on whether one considers the that rooftop solar does not require new land, the additional known global economically available reserves or the more spec- footprint and spacing areas required for all WWS power for all ulative potential global resource (Table 5). Thus, if Nd is to be used purposes worldwide would be only 0.41% and 0.59%, respec- beyond a few hundred years, it will have to be recycled from tively, of all land worldwide (or 1.0% of all land for footprint plus magnet scrap, a possibility that has been demonstrated (Takeda spacing). et al., 2006; Horikawa et al., 2006), albeit at unknown cost. However, even if the resource base and recycling could sustain high levels of Nd use indefinitely, it is not likely that actual global 5. Material resources production will be able to increase by a factor of five for many years, because of political or environmental limitations on expanding In a global all-WWS-power system, the new technologies supply (Lifton, 2009; Reisman, 2009). Therefore, it seems likely produced in the greatest abundance will be wind turbines, solar that a rapid global expansion of wind power will require many 1162 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169

Table 5 terawatt-level development of photovoltaics, the power production of Rare earth oxide and neodymium oxide (in parentheses)a production, reserves, and silicon PV technologies is limited not by crystalline silicon (because resources worldwide (million metric tonnes of rare earth oxide). silicon is widely abundant) but by reserves of silver, which is used as an Source: USGS (2009, p. 131). electrode (Feltrin and Freundlich, 2008). That review notes that ‘‘if the Country Mine production Reserves Reserve Base Resources use of silver as top electrode can be reduced in the future, there are no 2008 other significant limitations for c-Si solar cells’’ with respect to reaching multi-terawatt production levels (Feltrin and Freundlich, 2008, p. 182). United States 0 (0.000) 13 (2.0) 14 (2.1) n.r. For thin-film PVs, substituting ZnO electrodes for indium thin oxide Australia 0 (0.000) 5.2 (0.9) 5.8 (1.0) n.r. China 0.120 (0.022) 27 (4.9) 89 (16.0) n.r. allows multi-terawatt production, but thin-film technologies require CIS n.a. 19 (3.4) 21 (3.8) n.r. much more surface area. The limited availability of tellurium (Te) and India 0.003 (0.001) 1.1 (0.2) 1.3 (0.2) n.r. indium (In) reduces the prospects of cadmium telluride (CdTe) and Others 0.001 (0.000) 22 (4.0) 23 (4.1) copper indium gallium selenide (CIGS) thin cells. World total 0.124 (0.022) 88 (15.3) 150 (27.3) ‘‘very large’’b For multi-junction concentrator cells, the limiting material is CIS¼Commonwealth of Independent States. n.a.¼not available. ‘‘Reserves’’ are germanium (Ge), but substitution of more abundant gallium (Ga) ‘‘that part of the reserve base which could be economically extracted or produced at would allow terawatt expansion. the time of determination. The term reserves need not signify that extraction Wadia et al. (2009) estimate the annual electricity production facilities are in place and operative’’ (USGS, 2009, p. 192). The ‘‘Reserve Base’’ that would be provided by each of 23 different PV technologies if comprises reserves (as defined above), plus marginally economic resources, plus currently sub-economic resources. ‘‘Resources’’ comprise the reserve base (as either one year of total current global production or alternatively defined above) plus commodities that may be economically extractable in the the total economic reserves (as estimated by the USGS) of the future (USGS, 2009, p. 191). limiting material for each technology were used to make PVs. They a Assumes that the Nd oxide content of total rare earth oxides is 15% in the U.S. also estimate the minimum $/W cost of the materials for each of the and 18% in China, Australia, and all other countries (based on Table 2 of Hedrick, 23 PV technologies. They conclude that there is a ‘‘major oppor- 2009). tunity for fruitful new research and development based on low cost b The USGS (2009) writes that ‘‘undiscovered resources are thought to be very large relative to expected demand’’ (p. 131). and commonly available materials’’ (Wadia et al., 2009, p. 2076), such as FeS2, CuO, Cu2S, and Zn3P2. generators that do not use Nd (or other REE) PMs or a rapid On the basis of this limited review, we conclude that the transition into recycling. There are at least two kinds of development of a large global PV system is not likely to be limited alternatives: by the scarcity or cost of raw materials.

(i) generators that perform at least as well as PM generators but 5.3. Electric vehicles don’t have scarce REEs (e.g., switched-reluctance motors (Lovins and Howe, 1992)), new high-torque motors with For electric vehicles there are three materials that are of most inexpensive ferrite magnets, and possibly high-temperature concern: rare-earth elements (REEs) for electric motors, lithium for super-conducting generators (Hatch, 2009); lithium-ion batteries, and platinum for fuel cells. Some permanent- (ii) generators that don’t use REEs but have higher mass per unit of magnet ac motors, such as in the Toyota Prius hybrid power than do PM generators (the greater mass will require (Toyota, 2010), can use significant amounts of REEs. For example, the greater structural support if the generator is in the tower). motor in the Prius uses 0.2–1 kg of Nd or 3.6–16.7 kg/MW (Maximum EV, 2009; Gorman, 2009). The low estimate is based on the Morcos (2009) presents the most cogent summary of the assumption that the Prius’ permanent magnet motors are 55 kW, implications of any limitation in the supply of Nd for permanent with NDFeB magnet containing 31% Nd by mass (Maximum EV, magnets: 2009). The high kg/MW estimate assumes 60 kW motors (Toyota, 2010). Although this is an order of magnitude less than is used in A possible dwindling of the permanent magnet supply caused some wind-turbine generators (see discussion above), the total by the wind turbine market will be self-limiting for the potential demand for Nd in a worldwide fleet of BEVs with following reasons: large electric generators can employ a wide permanent-magnet motors would still be large enough to be of variety of magnetic circuit topologies, such as surface perma- concern. However, there are a number of electric motors that do not nent magnet, interior permanent magnet, wound field, use REEs, and at least one of these, the switched reluctance motor, switched reluctance, induction and combinations of any of currently under development for electric vehicles (e.g., Goto et al., the above. All of these designs employ large amounts of iron 2005), is economical, efficient, robust, and high-performing (Lovins (typically in the form of silicon steel) and copper wire, but not all and Howe, 1992). Given this, we do not expect that the scarcity of require permanent magnets. Electric generator manufacturers REEs will appreciably affect the development of electric vehicles. will pursue parallel design and development paths to hedge Next we consider lithium and platinum supply issues. To see against raw material pricing, with certain designs making the how lithium supply might affect the production and price of best economic sense depending upon the pricing of copper, steel battery-electric vehicles, we examine global lithium supplies, and permanent magnets. Considering the recent volatility of lithium prices, and lithium use in batteries for electric vehicles. sintered NdFeB pricing, there will be a strong economic Table 6 shows the most recent estimates of lithium production, motivation to develop generator designs either avoiding per- reserves, and resources from USGS (2009). manent magnets or using ferrite magnets with much lower and Note that Table 6 does not include the recently discovered, more stable pricing than NdFeB. potentially large lithium reserves in Afghanistan (Risen, 2010). Roughly half of the global lithium reserve base known in 2009 is in 5.2. Solar power one country, Bolivia, which has been called ‘‘the Saudi Arabia of lithium’’ (Friedman-Rudovsky, 2009). However, Bolivia does not Solar PVs use amorphous silicon, polycrystalline silicon, micro- yet have any economically recoverable reserves or lithium produc- crystalline silicon, cadmium telluride, copper indium selenide/sulfide, tion infrastructure (Ritter, 2009; Wright, 2010), and to date has not and other materials. According to a recent review of materials issues for produced any lithium (Table 6). About 75% of the world’s known M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1163

Table 6 Ultimately, then, the issue of how the supply of lithium affects Lithium production, reserve,s and resources worldwide as of 2009 (metric tonnes). the viability of lithium-ion-battery EVs in an all-WWS world boils Source: USGS 2009. down to the price of lithium with sustainable recycling. As noted

Country Mine production Reserves Reserve Base Resources above, it does make some difference to EV economics if that price is 2008 $35/kg-Li or $100/kg-Li. Finally we consider the use of platinum in fuel cells. The United States n.r. 38,000 410,000 n.r. production of millions of hydrogen fuel cell vehicles (HFCVs) Argentina 3200 n.r. n.r. n.r. would increase demand for Pt substantially. Indeed, the production Australia 6900 170,000 220,000 n.r. Bolivia 0 0 5,400,000a n.r. of 20 million 50 kW HFCVs annually might require on the order of Chile 12,000 3,000,000 3,000,000 n.r. 250,000 kg of Pt—more than the total current world annual China 3500 540,000 1,100,000 n.r. production of Pt (Yang, 2009; USGS, 2009, p. 123). How long this World total 27,400 4,100,000 11,000,000 413,000,000 output can be sustained, and at what platinum prices, depends n.r.¼not reported. For explanation of terms, see notes to Table 5. on several factors: (1) the technological, economic, and institu- tional ability of the major supply countries to respond to changes a Wright (2010, p. 58) reports that the head of the Bolivian scientific committee charged with developing Bolivia’s lithium resources estimates that there are about in demand; (2) the ratio of recoverable reserves to total production; 100,000,000 metric tonnes of metallic lithium in Bolivia. (3) improvements in technology that reduce the cost of recovery; and (4) the cost of recycling as a function of quantity recycled. economically recoverable reserves are in Chile, which is also the Regarding the first factor, it does not seem likely that the current world’s leading producer (Table 6). Both Bolivia and Chile recognize production problems in South Africa, mentioned by Yang (2009), the importance of lithium to battery and car makers, and are hoping will be permanent. Rather, it seems reasonable to assume that in to extract as much value from it as possible (Wright, 2010). This the long run, output can be increased in response to large changes concentration of lithium in a few countries, combined with rapidly in demand and price. In support of this, the U.K. Department of growing demand, could increase the price of lithium upon Transport (UKDOT, 2006) cites a study that concludes that expanded BEV production. Currently, lithium carbonate (Li2CO3) ‘‘production in South Africa could be expanded at a rate of 5% costs $6–7/kg, and lithium hydroxide (LiOH), $10/kg (Jaskula, per year for at least another 50 years’’. TIAX (2030) finds that ‘‘the 2008), which correspond to about $35/kg Li. Lithium is 1–2% of platinum industry has the potential to meet a scenario where FCVs the mass of a lithium-ion battery (Gaines and Nelson, 2009; achieve 50% market penetration by 2050, while an 80% scenario Wilburn, 2009, Table A-9); in a pure BEV with a relatively long could exceed the expansion capabilities of the industry’’ (p. 7). range (about 100 miles), the battery might contain on the order of Regarding the second factor, Spiegel (2004) writes that the 10 kg of lithium (Gaines and Nelson, 2009). At current prices this International Platinum Association concludes that ‘‘there are adds $350 to the manufacturing cost of a vehicle battery, but if sufficient available reserves to increase supplies by up to 5–6% lithium prices were to double or triple, the lithium raw material per year for the next 50 years,’’ (p. 364), but does not indicate what cost could approach $1000, which would increase vehicle costs the impact on prices might be. Gordon et al. (2006) estimate that 29 further. million kg of platinum-group metals are available for future use, At 10 kg per vehicle, the production of 26 million EVs per year – and state that ‘‘geologists consider it unlikely that significant new more than half of the 48 million passenger cars produced in the platinum resources will be found’’ (p. 1213). This will sustain world in 2009 (OICA, 2010) – would require 260,000 metric tonnes- annual production of at least 20 million HFCVs, plus production of Li per year, which in the absence of recycling lithium batteries conventional catalyst-equipped vehicles, plus all other current (which currently is negligible) would exhaust the current reserve non-automotive uses, for less than 100 years, without any base (Table 6) in less than 50 years. If one considers an even larger recycling. EV share of a growing, future world car market, and includes other Regarding the third factor, TIAX (2003) argues that in the long demands for lithium, it is likely that the current reserve base would run the price of platinum is stable because the extra cost of be exhausted in less than 20 years, in the absence of recycling. This recovering deeper and more diffuse reserves is balanced by is the conclusion of the recent analysis by Meridian International technological improvements that reduce recovery costs. It is not Research (2008). clear, however, that this improvement can be expected to continue However, the world will not consume lithium reserves in an indefinitely. Thus, the prospects for very long term use of platinum, uncontrolled manner until, one day, the supply of lithium is and the long-term price behavior of platinum, depend in large part exhausted. As demand grows the price will rise and this will spur on the prospects for recycling (TIAX, 2003). the hunt for new sources of lithium, most likely from recycling. According to an expert in the precious-metal recycling industry, Another potential source of lithium is the oceans, which contain the full cost of recycled platinum in a large-scale, international 240 million tonnes, far more than all the known land reserves. recycling system is likely to be much less than the cost of producing However, currently the cost of extracting such lithium is high and virgin platinum metal (Hageluken,¨ 2009). Consistent with this, energy intensive, so alternatives are strongly preferred. According UKDOT (2006) cites an analysis that indicates that platinum to an expert, recycling lithium currently is more expensive than is recycling will be economical even if platinum loadings on fuel- mining virgin material (Ritter, 2009), but as the price of lithium cell catalysts are greatly reduced from current levels. Thus, the rises, at some point recycling will become economical. The more the recycling, the less the production of high-cost virgin economics of recycling depend in part on the extent to which material, and hence the lower the price of platinum, since the price batteries are made with recyclability in mind, an issue that the will be equal to the long-run marginal cost of producing virgin major industries already are aware of: according to a recent report, metal. The effect of recycling on platinum price, therefore, depends ‘‘lithium mining companies, battery producers, and automakers on the extent of recycling. have been working together to thoroughly analyze lithium avail- The prospects for recycling are difficult to quantify, because ability and future recyclability before adopting new lithium-ion they depend more on institutional and logistical factors than on chemistries’’ (Ritter, 2009, p. 5). Gaines and Nelson (2010) discuss technical factors. The current rate of recycling autocatalysts is recycling processes for lithium-ion batteries, and write that between 10% and 25%, if expressed as the ratio {Pt recovered from ‘‘recovery of battery-grade material has been demonstrated’’ (p. 7). catalysts in year X}:{Pt used in new catalysts in year X} (Carlson and 1164 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169

Thijssen, 2002; Hageluken¨ et al., 2009; Hageluken,¨ 2009), but is One scenario for powering the world with a WWS system around 50% if expressed as the ratio {Pt recovered from catalysts in includes 3.8 million 5 MW wind turbines (supplying 50% of year X}:{Pt used in new catalysts in the year in which the currently projected total global power demand in 2030), 49,000 300 MW recycled products were made} (Hageluken,¨ 2009 (also quoted in CSP power plants (supplying 20% of demand), 40,000 solar PV Ritter, 2009, p. 4)). This second ratio, representing the ‘‘dynamic power plants (14%), 1.7 3 kW rooftop PV systems (6%), 5350 recycling rate,’’ is more meaningful because it is based on the 100 MW geothermal power plants (4%), 900 1300 MW hydro- lifecycle of a particular product (e.g., Schaik and Reuter, 2004). electric power plants, of which 70% are already in place (4%), Technically, there appears to be ample room to increase dynamic 720,000 0.75 MW wave devices (1%), and 490,000 1 MW tidal recycling rates. Hageluken¨ et al. (2009) believe that ‘‘a progressive turbines (1%). conversion of existing open loop recycling systems to more The equivalent footprint area on the ground for the sum of WWS efficient closed loopsywould more than double the recovery of devices needed to power the world is 0.74% of global land area; PGMs from used autocatalysts by 2020’’ (p. 342). (Hageluken¨ et al. the spacing area is 1.16% of global land area. Spacing area can be (2009) and UKDOT (2006) also note that emissions from recycling used for multiple purposes, including agriculture, ranching, and PGMs are significantly lower than emissions from mine production open space. However, if one-half of the wind devices are placed of PGMs.) Spiegel (2004) states that ‘‘technology exists to profitably over water, if we consider wave and tidal devices are in water, and if recover 90% of the platinum from catalytic converters’’ (p. 360), and we consider that 70% of hydroelectric is already developed and in his own analysis of the impact of HFCV platinum on world rooftop solar areas are already developed, the additional footprint platinum production, he assumes that 98% of the Pt in HFCVs will be and spacing of devices on land required are only 0.41% and recoverable. Similarly, Hageluken,¨ 2009 asserts that the technology 0.59% of the world land area, respectively. is available to recover more than 90% of the platinum from fuel The development of WWS power systems is not likely to be cells, although he believes that 98% recovery will be difficult to constrained by the availability of bulk materials, such as steel and achieve. Finally, in their separate analyses of the impact of the concrete. In a global WWS system, some of the rarer materials, such introduction of hydrogen HFCVs on platinum supply and prices, as neodymium (in electric motors and generators), platinum UKDOT (2006) and TIAX (2003) assume that 95% of the platinum in (in fuel cells), and lithium (in batteries), will have to be recycled fuel cells will be recovered and recycled. (UKDOT (2006) cites two or eventually replaced with less-scarce materials unless additional sources, one of them a catalyst manufacturer, in support of its resources are located. The cost of recycling or replacing neody- assumption.) mium or platinum is not likely to affect noticeably the economics of It seems likely that a 90%+ recycling rate will keep platinum WWS systems, but the cost of large-scale recycling of lithium prices lower than will a 50% recycling rate. The main barriers to batteries is unknown. achieving a 90%+ recycling rate are institutional rather than In Part II of this study (Delucchi and Jacobson, this issue), we technical or economic: a global recycling system requires inter- examine reliability, system and transmission costs, and policies national agreement on standards, protocols, infrastructure, man- needed for a worldwide WWS infrastructure. agement, and enforcement (Hageluken,¨ 2009). We cannot predict when and to what extent a successful system will be developed. Nevertheless, it seems reasonable to assume that enough Acknowledgements platinum will be recycled to supply a large and continuous fuel-cell vehicle market with only moderate increases in the price We would like to thank Elaine Hart and Dave Elliott for helpful of platinum, until new, less costly, more abundant catalysts or comments. This paper was not funded by any interest group, fuel cell technologies are found. Indeed, catalysts based on company, or government agency. inexpensive, abundant non-platinum materials may be available soon (e.g., Lefevre et al., 2009). Preliminary work by Sun et al. (2010) supports this conclusion. They developed an integrated Appendix A model of HFCV production, platinum loading per HFCV (a function of HFCV production), platinum demand (a function of HFCV A.1. The economics of nuclear power production, platinum loading, and other factors), and platinum prices (a function of platinum demand and recycling), and found The economics of nuclear power are discussed in Kessides that in a scenario in which HFCV production was increased to 40% (2010), Grubler (2010), Joskow and Parsons (2009), Feiveson of new LDV output globally in the year 2050, the average platinum (2009), Koomey and Hultman (2007), Hultman et al. (2007), cost per HFCV was $400, or about 10% of the cost of the fuel-cell Hultman and Koomey (2007), Harding (2007), and Deutch et al. system. (2003, 2009). Kessides (2010) and Joskow and Parsons (2009) discuss at length the issues that affect the economics of nuclear power. Feiveson (2009) reviews recent escalation in capital costs. Grubler (2010) argues that the real costs of nuclear power can 6. Summary of technical findings and conclusions increase with an expansion of capacity (and in fact did increase in France) because of ever-increasing complexity in the design, This is Part I of a study to examine the feasibility of providing all construction, operation, management, and regulatory oversight energy for all purposes (electric power, transportation, heating/ of nuclear systems. Koomey and Hultman (2007) estimate that the cooling, etc.) worldwide from wind, water, and the sun (WWS). The total levelized busbar costs of 99 US reactors, including capital costs main technical findings of this analysis are as follows: amortized at 6%/year, range from $0.03/kWh to $0.14/kWh (2004 Converting to a WWS energy infrastructure will reduce 2030 USD), with the 50% percentile falling between $0.05/kWh and world power demand by 30%, primarily due to the efficiency of $0.06/kWh. Hultman et al. (2007) argue that costs at the upper end electricity compared with internal combustion. The amount of of the $0.03–0.14/kWh range are driven in part by unanticipated wind power plus solar power available in likely developable factors, and Hultman and Koomey (2007) argue that the possibility locations over land outside of Antarctica worldwide to power of such ‘‘cost surprises’’ should be incorporated formally into cost the world for all purposes exceeds projected world power demand estimates for nuclear power. Koomey and Hultman (2007) argue by more than an order of magnitude. that standardization of design, improvements in construction M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1165 management, computer-assisted design, and other factors might A.2.4. Commercial sector tend to drive costs down, but that the special conditions that attend We assume that the fraction of energy-end use that can be each nuclear job site, and the possibility of cost ‘‘surprises,’’ tend to electrified is slightly less than we estimated for the residential drive costs up. Deutch et al. (2003) estimate that the real levelized sector, except in the case of renewables. cost of nuclear power using an ‘‘open’’ or ‘‘once-through’’ fuel cycle (in which spent fuel is treated as waste, rather than recycled back to the reactor) ranges from $0.04 to $0.08/kWh (2002 USD) (with an A.2.5. Industrial sector effective interest rate of 11.5%), depending on assumptions regard- We assume that 50% of direct-process heat end use, 50% of ing the , the plant lifetime, construction costs, and and combined heat-and-power end use, and 25% of construction time. Deutch et al. (2009) and Du and Parsons (2009) conventional boiler fuel use are not electrified, and then use data on estimate that since the Deutch et al. (2003) report, construction manufacturing consumption of energy in the U.S. (Table 2.3 of EIA costs have escalated substantially, resulting in a doubling of capital (2008b)) to calculate a weighted-average electrified fraction by costs and an increase in the estimated median levelized cost from type of fuel. We assume that the estimates calculated on the basis of $0.067/kWh in the 2003 study (2002 USD) to $0.084/kWh in the U.S. data apply to the world. 2009 update (2007 USD) (see also Joskow and Parsons, 2009). Harding (2007) estimates even higher levelized costs of A.2.6. Transport sector $0.09–0.12/kWh (2007 USD). We assume that 5% of motor-gasoline use, 30% of highway In summary, the costs of nuclear power are estimated to cover a diesel-fuel use, 50% of off-road use, 100% of military fuel very wide range, depending on a number of variables that are use, 20% of train fuel use, and 100% of airplane and ship fuel use are difficult to project: the costs of new, untested designs; construction not electrified. We use data on transport energy consumption times; interest rates; the impact of unforeseen events; regulatory from the IEA (2008, p. 464, 508), data on transport fuel use in the requirements; the potential for economies of scale; site- and U.S. (EIA, 2008b, Table 5.14c), and data on diesel fuel use in the U.S. job-specific design and construction requirements; the availability (EIA, 2008b, Table 5.15) to estimate a weighted-average electrified of specialty labor and materials; bottlenecks in the supply chains; fraction by type of fuel. We assume that estimates calculated on the the potential for standardization; and so on. basis of U.S. data apply to the world.

A.2. Notes to Table 2 A.2.7. Non-electrified energy services A.2.1. TW power in 2030 (fossil-fuel case) We assume that the remaining (non-electrified) energy service This is the projected total world and total U.S. power for all demands are met by hydrogen derived from electrolysis of water energy end uses in the year 2030, in the conventional or business- using WWS power. For analytical simplicity we assume that WWS as-usual scenario relying primarily on fossil fuels. The projections power is delivered to the site of hydrogen use or refueling and are from EIA (2008a); we converted from BTUs per year to Watts. used there to produce hydrogen electrolytically. (This is a useful The breakdown here is by type of energy in end use; thus, simplification because it obviates the need to analyze a hydrogen ‘‘renewables’’ here refers, for example, to end-use combustion of transmission system.) We assume that in all sectors except biomass, such as wood used for heating. transportation (e.g., in many industrial processes) the electrolyti- cally produced hydrogen is burned directly to provide heat. In the transportation sector except aviation, we assume that hydrogen is A.2.2. Electrified fraction compressed and then used in a fuel cell. This is the fraction of energy service demand in each sector that For aviation, we assume that hydrogen is liquefied and burned in jet can be satisfied feasibly by direct electric power. For example, gas engines. Coenen (2009), Nojoumi et al. (2009), Janic (2008), Maniaci water heating and space heating can readily be converted to (2006), Mital et al. (2006), Corchero and Montan˜es (2005), Koroneos air- and ground-source heat-pump water heaters and air heaters et al. (2005),andWestenberger (2003) discuss various aspects of and electric resistance heaters. Liquid-fuel internal-combustion- liquid-hydrogen-powered aircraft. Westenberger (2003), reporting on engine vehicles can be replaced by battery electric vehicles. Indeed, a European analysis of liquid-hydrogen aircraft systems (the direct electricity can, technically, provide almost any energy CRYOPLANE project), concludes that hydrogen is a ‘‘suitable alternative service that fuel combustion can, with the likely exception of fuel for future aviation’’ (p. 2), and could be implemented within 15–20 transportation by air. However, in other cases, even if it is years (of 2003) with continued research and development of engines, technically feasible, it may be relatively expensive or difficult for materials, storage, and other components. Corchero and Montan˜es electricity to provide exactly the same service that fuel combustion (2005) also discuss the CRYOPLANE project and conclude that ‘‘evolving does: for example, some cooking and heating applications where a a conventional engine from burning kerosene to burning hydrogen, flame is preferred, some large-scale direct uses of process heat, without implementing large-scale hardware changes, does not seem to some applications of combined heat and power production, and be an insurmountable task’’ (p. 42). Whereas, liquefied hydrogen some forms of heavy freight transportation. As explained below, we aircraft would require about four times more volume to store their fuel, assume that energy services that are not electrified are provided by they would require three times less mass, since hydrogen is combustion of electrolytic hydrogen. Our assumptions regarding one-twelfth the density of jet fuel. Coenen (2009) asserts that ‘‘LH2 the directly electrified fraction in each sector are as follows: fueled aircraft are lighter, cleaner, quieter, safer, more efficient and have greater payload and range for equivalent weight of Jet A fuel,’’ and that A.2.3. Residential sector ‘‘there are no critical technical barriers to LH2 air transport’’ (p. 8452). We assume that 5% of fuel use for space heating and 20% of fuel use Koroneos et al. (2005) perform a lifecycle assessment of the environ- for ‘‘appliances’’ (mainly cooking) are not electrified, and then use the mental impacts of jet fuel and hydrogen made from various feedstocks, data from Table 2.5 of EIA (2008b) to calculate a weighted-average and find that hydrogen made from water and wind power has the electrifiable fraction by type of fuel. We assume that renewables are lowest impacts across all dimensions. For a discussion of liquid jet fuels mainly fuelwood, which will not be replaced with electricity. We made from biomass, see Hileman et al. (2009). assume that the estimates calculated on the basis of U.S. data apply to Thus, in transportation, all vehicles, ships, trains, and planes are the world. either battery-powered or hydrogen powered. In this way, WWS 1166 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 power meets all energy needs, either directly as electricity or hydrogen compression or liquefaction. Our estimation of these indirectly via electrolytic hydrogen. factors is based on the data in Table A1. Although 5–10% of the volumetric output of refineries is non- fuel product such as lubricants, petrochemical feedstocks, A.2.8. End-use energy/work w.r.t. to fossil fuel road asphalt, and petroleum coke (EIA, 2010), these products This is the ratio of BTUs-electric/unit-work to BTUs-fossil-fuel/ require much less than 5–10% of refinery energy, because refinery unit-work. For example, it is the ratio of BTUs of electricity (at 3412 energy is used disproportionately to produce highly refined BTUs/kWh) input to an electric vehicle from the outlet, per mile of transportation fuels (Delucchi, 2003). Moreover, some of these travel provided, to BTUs of gasoline input to a conventional vehicle non-fuel products would be eliminated in a WWS world (e.g., some from the pump, per mile of travel provided. In the case of electrified kinds of lubricants), and some could be replaced at very low energy end uses, BTUs-electric are measured at the point of end use, and do cost, for example, by recycling. For these reasons, we do not not include any upstream or ‘‘indirect’’ electricity uses. In the attempt to estimate the very small amount of refinery energy case of electrolytic hydrogen (eH2), BTUs-electric are measured (probably on the order of 2%) that still would be required in a at the input to the electrolyzer, which for simplicity is assumed to WWS world. be at the site of end use, and again do not include any upstream or indirect electricity uses such as for hydrogen compression. (We treat compression and liquefaction separately, in the

‘‘upstream factor’’ column.) Thus, the figures shown for eH2 include A.2.10. EHCM factor losses during electrolysis. Our estimates are based on results or EHCM stands for ‘‘electricity and hydrogen conservation mea- assumptions from the Advanced Vehicle Cost and Energy Use Model sure.’’ This is the ratio of demand for end-use energy after EHCMs (AVCEM) (Delucchi, 2005), the Lifecycle Emissions Model (LEM) have been instituted to the demand for end-use energy before the (Delucchi, 2003), and other sources, as listed in Table A1. EHCMs. Demand-side energy-conservation measures include improving the energy-out/energy-in efficiency of end uses (e.g., with more efficient vehicles, more efficient lighting, better A.2.9. Upstream factor insulation in homes, and the use of heat-exchange and filtration The upstream factor accounts for changes, in a WWS world systems), directing demand to low-energy-use modes (e.g., using compared with the base-case fossil-fuel world, in sectoral energy public transit or telecommuting in place of driving), large-scale use in activities that are ‘‘upstream’’ of final end use by consumers. planning to reduce overall energy demand without compromising We first discuss these changes qualitatively, and then provide economic activity or comfort (e.g., designing cities to facilitate quantitative estimates of the changes in upstream fuel processing greater use of non-motorized transport and to have better match- activities, which we believe are the largest of the upstream ing of origins and destinations (thereby reducing the need for changes. travel)), and designing buildings to use solar energy directly (e.g., In a WWS world some of the energy-generation technologies with more daylighting, solar hot water heating, and improved (such as wind turbines), forms of energy (such as compressed passive solar heating in winter and cooling in summer). We assume hydrogen), and energy-use technologies (such as electric vehicles) that EHCMs can achieve modest reductions in energy demand, on will be different from those in a conventional fossil-fuel world. the order of 5–15% in most cases. These differences will give rise to differences in energy use in the sectors that manufacture energy technologies and process energy. Qualitatively these differences are described in Table A2. Table 2 has upstream adjustment factors for fuel use in the A.2.11. TW power in 2030 (WWS case) industrial sector and liquid fuel use in the transportation sector. These are the world and the U.S. power in the year 2030 when The factors shown in Table 2 for the industrial sector account for the wind, water, and solar power provide all energy services, and thus elimination of energy use in petroleum refining. The factor shown replace 100% of fossil-fuel use and biomass combustion. It is for liquid fuel in transportation accounts for electricity use for calculated from the other values in the table.

Table A1 Parameters values used to derive results in Table 2.

Value Parameter Data source

0.80 Efficiency of fossil-fuel heating (BTUs-work/BTUs-input-energy) LEM (Delucchi, 2003) 0.97 Efficiency of electric resistance heating (BTUs-work/BTUs-power) LEM (Delucchi, 2003) 0.80 Efficiency of hydrogen heating (BTUs-work/BTUs-input-energy) Assume same as fossil fuel

0.70 Efficiency of electrolytic hydrogen production on site (BTUs-H2/BTUs-electricity, AVCEM, LEM (Delucchi, 2003, 2005; Aguado et al., 2009, higher heating value) assume 75%) 1.10 Work/energy ratio of hydrogen combustion in engines (mainly jet engines) relative to LH2 in vehicles is more efficient than gasoline ratio for petroleum fuel

0.15 Of total liquid fuel use in transportation, the fraction that is replaced with liquefied H2 Assume LH2 used by airplanes and some ships (EIA, 2008b,

rather than compressed H2, on an energy basis Table 5.14c) 5.30 Ratio of mi/BTU for EVs to mi/BTU ICEVs AVCEM (Delucchi, 2005) 2.70 Ratio of mi/BTU for HFCVs to mi/BTU ICEVs AVCEM (Delucchi, 2005)

1.12 Multiplier for electricity requirements of H2 compression for transportation AVCEM (Delucchi, 2005)

(10,000 psi) (BTUs-electricity plus BTUs-H2/BTU-H2)

1.32 Multiplier for electricity requirements of H2 liquefaction for transportation, mainly air AVCEM (Delucchi, 2005)

transport (includes boil-off losses) (BTUs-electricity plus BTUs-H2/BTU-H2) 0.28 Petroleum energy in oil refining as a fraction of total petroleum use in industrial sector Projections for the U.S. for the year 2030 (EIA, 2009, Table 6) 0.18 NG energy in oil refining as a fraction of total NG use in industrial sector Projections for the U.S. for the year 2030 (EIA, 2009, Table 6) 0.27 Coal energy in oil refining as a fraction of total coal use in industrial sector Projections for the U.S. for the year 2030 (EIA, 2009, Table 6) 0.07 Electricity in oil refining as a fraction of total electricity use in industrial sector Projections for the U.S. for the year 2030 (EIA, 2009, Table 6) M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1167

Table A2 Qualitative differences between a fossil-fuel and a WWS world.

Sector Fossil-fuel world WWS world Difference Our treatment

Mining—oil, gas, and Energy use in this sector typically Energy use for mining for non- Small reduction in energy use in a Not estimated. coal is 1–4% of final fuel energy fuel products only. WWS world. (Delucchi, 2003).

Mining—other Energy use in mining of metals Energy use in mining of metals A WWS world will require more Not estimated. metals and and minerals for the production and minerals for the production bulk materials (e.g., steel, minerals of power plants, motors, engines, of power plants, motors, engines, concrete, etc.) per unit of power vehicles, equipment, etc.; vehicles, equipment, etc. Because output than does a fossil-fuel probably very small fraction of the types of equipment (etc.) world, and hence probably will total energy use. produced will be different than in require more energy in mining the fossil-fuel world, mining raw materials. This increase in energy use will be different. energy use likely is small.

Manufacturing—ma- Energy use to make and assemble Energy use to make and assemble To the extent that WWS Not estimated. terials production finished materials for power finished materials for power technologies have greater mass and assembly for plants, motors, engines, vehicles, plants, motors, engines, vehicles, per unit of power output (e.g., energy generation- equipment, etc. For reference, equipment, etc. Because the battery-electric vehicles versus and use- direct and indirect energy in the types of equipment (etc.) gasoline vehicles), the technologies manufacture of motor vehicles is produced will be different than in manufacturing energy use will be 5–15% of the lifetime fuel energy the fossil-fuel world, greater. (Delucchi, 2003). manufacturing energy use will be different.

Manufacturing—ma- Energy use to make pipelines, Energy use to make pipelines, A WWS world will not have Not estimated. terials production tankers, trucks, trains, and power tankers, trucks, trains, and power pipelines, tankers, trucks, or and assembly for lines that carry energy, energy lines that carry energy, energy trains delivering fuel or fuel energy delivery feedstocks, vehicles, and feedstocks, vehicles, and feedstocks, but will have more infrastructure materials for the energy system. materials for the energy system. power lines, on balance, small reduction in energy use in WWS world?

Manufacturing—fuel Petroleum refining and natural- Compression and liquefaction of Significant net energy reduction Estimated; see Table 2 process gas processing. hydrogen. in WWS world. and Section A.2.9.

References Corchero, G., Montan˜es, J.L., 2005. An Approach to the use of hydrogen for commercial aircraft engines. Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering 219, 35–44. Adamantiades, A., Kessides, I., 2009. Nuclear power for sustainable development: Cullen, J.M., Allwood, J.M., 2009. The efficient use of energy: Tracing the global current status and future prospects. Energy Policy 37, 5149–5166. flow of energy from fuel to service. Energy Policy doi:10.1016/j.enpol.2009. Aguado, M., Ayerbe, E., Azca´rate, C., Blanco, R., Garde, R., Mallor, F., Rivas, D.M., 2009. 08.054. Economical Assessment of a wind–hydrogen energy system using WindHyGen Cryosphere Today, Current Northern Hemisphere Sea Ice Area accessed August 30, 2010. software. International Journal of Hydrogen Energy 34, 2845–2854. /http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.recent.arctic.png.S. Alliance for Climate Protection, Repower America, 2009. /http://www.repower Czisch, G., 2006. Low cost but totally renewable electricity supply for a huge supply america.org/S. area—a European/Trans-European example. Unpublished manuscript . /http:// Anderson, L.G., 2009. use in Brazil: air quality impacts. Energy and www.iset.uni-kassel.de/abt/w3-w/projekte/LowCostEuropElSup_revised_for_ Environmental Science 2, 1015–1037. AKE_2006.pdfS. American Physical Society, 2008. Energy future: think efficiency, how America can Delucchi, M.A., 2010. Impacts of Biofuels on Climate, Land, and Water. Annals of the look within to achieve and reduce global warming. /http:// New York Academy of Sciences 1195, 28–45 (Issue The Year in www.aps.org/energyefficiencyreport/report/aps-energyreport.pdfS. Ecology and Conservation Biology, Ostfeld, R.S. and W. H. (Eds.). Schlesinger.). Archer, C.L., Jacobson, M.Z., 2005. Evaluation of global windpower. Journal of Delucchi, M.A., 2003. A lifecycle emissions model (LEM): lifecycle emissions from Geophysics Research. Barre´, B., 1999. ‘‘Nuclear energy’’ From here to where? Nuclear Physics A654, transportation fuels, motor vehicles, transportation modes, electricity use, 409c–416c. heating and cooking fuels, and materials. Main Report, UCD-ITS-03-17. Institute / Beyond Zero Emissions, 2010. Zero carbon Australia stationary energy plan, of Transportation Studies, University of California, Davis. http://www.its. S Melbourne Energy Institute, University of Melbourne. /http://beyondzeroemis ucdavis.edu/people/faculty/delucchi/ . sions.org/S. Delucchi, M.A., 2005. Overview of AVCEM. UCD-ITS-RR-05-17 (1). Institute of / California Fuel Cell Partnership, 2009. Hydrogen fuel-cell vehicle and station Transportation Studies, University of California, Davis. http://www.its.ucda S deployment plan: a strategy for meeting the challenge ahead. West Sacramento, vis.edu/people/faculty/delucchi/ . California. /http://www.cafcp.org/sites/files/Action_Plan_Final.pdfS. Delucchi, M.Z., Jacobson, M.Z., (this issue). Providing all global energy with wind, Carlson, E.J., Thijssen, J.H.J. , 2002. Precious metal availability and cost analysis for water, and solar power, part II: reliability, system and transmission costs, and PEMFC commercialization, part IV.F.1. (pp. 513–516) of hydrogen, fuel cells and policies. Energy Policy, doi:10.1016/j.enpol.2010.11.045. infrastructure technologies. FY 2002 Progress Report, U.S. Department of Deutch, J., et al., 2009. Update of the MIT 2003 Future of Nuclear Power. Energy, Energy Efficiency and Renewable Energy, Office of Hydrogen, Fuel Cells, Massachusetts Institute of Technology, Cambridge, Massachusetts. /http:// and Infrastructure Technologies, Washington, D.C. web.mit.edu/nuclearpower/S. Chow, T.T., 2010. A review of photovoltaic/thermal hybrid solar technology. Applied Deutch, J., et al., 2003. The Future of Nuclear Power. Massachusetts Institute of Energy 87, 365–379. Technology, Cambridge, Massachusetts. /http://web.mit.edu/nuclearpower/S. Cleetus, R., Clemmer, S., Friedman, D., 2009. Climate 2030: a national blueprint for a Du, Y., J.E. Parsons, 2009. Update on the Cost of Nuclear Power, 09-004. Center for clean energy economy, Union of Concerned Scientists, Cambridge, Massachusetts. Energy and Environmental Policy Research, Massachusetts Institute of Tech- /http://www.ucsusa.org/global_warming/solutions/big_picture_solutions/clima nology, Cambridge, Massachusetts. /http://web.mit.edu/ceepr/www/publica te-2030-blueprint.html#Download_the_Climate_2030_Blueprint_repo.S. tions/workingpapers/2009-004.pdfS. Coenen, R.M., 2009. A proposal to convert air transport to clean hydrogen (CATCH). Dvorak, M., Archer, C.L., Jacobson, M.Z., 2010. California offshore wind energy International Journal of Hydrogen Energy 34, 8451–8453. potential. Renewable Energy 35, 1244–1254 doi:10.1016/j.renene.2009.11.022. Colby, D.W., et al., 2009. Wind turbine sound and health effects, an expert review Energy Information Administration (EIA), 2008a. International Energy Outlook panel, American Wind Energy Association. /http://awea.org/newsroom/ 2008, DOE/EIA-0484(2008). U.S. Department of Energy, Washington, D.C. releases/AWEA_CanWEA_SoundWhitePaper_12-11-09.pdf.S. /http://www.eia.doe.gov/oiaf/ieo/index.htmlS. 1168 M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169

Energy Information Administration (EIA), 2008b. Annual Energy Review 2007, DOE/ Hileman, J.I., Ortiz, D.S., Bartis, J.T., Wong, H.M., Donohoo, P.E., Weiss, M.A., Waitz, EIA-0384(2007). U.S. Department of Energy, Washington, D.C., June 23, 2008. I.A., 2009. Near-term feasibility of alternative jet fuels. RAND TR554, Rand /http://www.eia.doe.gov/emeu/aer/consump.htmlS. Corporation and Massachusetts Institute of Technology. /http://www.rand. Energy Information Administration (EIA), 2009. Annual Energy Outlook 2009, DOE/ org/pubs/technical_reports/2009/RAND_TR554.pdfS. EIA-0383(2009). U.S. Department of Energy, Washington, D.C. /http://www. Hoffert, M.I., et al., 2002. Advanced technology paths to global climate stability: eia.doe.gov/oiaf/aeo/index.htmlS. energy for a greenhouse planet. Science 298, 981–987. Energy Information Administration (EIA), Petroleum Navigator, July 29, 2010 Hultman, N.E., Koomey, J.G., Kammen, D.M., 2007. What history can teach us about /http://tonto.eia.doe.gov/dnav/pet/PET_PNP_PCT_DC_NUS_PCT_A.htmS (last the future costs of U.S. nuclear power. Environmental Science and Technology 1, accessed August 30, 2010). 288–293. Energy Net Nuclear: Diablo Canyon History II. http://www.energy-net.org/01NUKE/ Hultman, N.E., Koomey, J.G., 2007. The risk of surprise in energy technology costs. DIABLO2.HTM (last accessed August 28, 2010). Environmental Research Letters 2, 034002 doi:10.1088/1748-9326/2/3/ European Climate Foundation (ECF), 2010. Roadmap 2050: a practical guide to a 034002. prosperous, low-carbon Europe. /http://www.europeanclimate.org/S. Hurst, C., 2010. China’s rare earth elements industry: what can the west learn? European Renewable Energy Council (EREC), 2010. Re-thinking 2050: a 100% Institute for the Analysis of Global Security, Washington, D.C. /http://fmso. renewable energy vision for the European Union. /http://www.erec.orgS. leavenworth.army.mil/documents/rareearth.pdfS. Federation of American Scientists, 2010. Nuclear weapons. http:/www.fas.org/ Intergovernmental Panel on Climate Change (IPCC), 2005. IPCC special report on nuke/guide/india/nuke/ (last accessed August 28, 2010). carbon dioxide capture and storage. Prepared by working group III. In: Metz, B., Feiveson, H.A., 2009. A skeptics view of nuclear energy. Daedalus, Fall, 60–70. Davidson, O., de Coninck, H.C., Loos, M. and Meyer, L.A. (Eds.). Cambridge Feltrin, A., Freundlich, A., 2008. Material considerations for terawatt level devel- University Press, Cambridge, United Kingdom and New York, NY, USA, 442 pp. opment of photovoltaics. Renewable Energy 33, 180–185. /http://arch.rivm.nl/env/int/ipcc/S. Friedman-Rudovsky, J., 2009. For lithium car batteries, Bolivia is in the driver’s seat. International Energy Agency (IEA), 2008. World Energy Outlook 2008, Organization Time Magazine, January 22 /http://www.time.com/time/world/article/ for Economic Cooperation and Development, Paris, France. 0,8599,1872561,00.htmlS. Jacobson, M.Z., 2007. Effects of ethanol () versus gasoline vehicles on cancer and Fridleifsson, I.B., Bertani, R., Huenges, E., Lund, J.W., Ragnarsson, A., Rybach, L., 2008. mortality in the United States. Environmental Science and TechnologyI 41, The possible role and contribution of geothermal energy to the mitigation of 4150–4157. doi: 10.1021/es062085v. climate change. In: O. Hohmeyer and T. Trittin (Eds.), IPCC Scoping Meeting on Jacobson, M.Z., 2009. Review of solutions to global warming, air pollution, and Renewable Energy Sources, Proceedings, Luebeck, Germany, 20–25 January energy security. Energy and Environmental Science 2, 148–173. doi:10.1039/ 2010, pp. 59–80. b809990c. Fthenakis, V., Mason, J.E., Zweibel, K., 2009. The technical, geographical, and Jacobson, M.Z., Masters, G.M., 2001. Exploiting wind versus coal. Science 293, 1438. economic feasibility of solar energy to supply the energy needs of the US. Jacobson, M.Z., M.A. Delucchi, 2009. A path to sustainable energy by 2030, Scientific Energy Policy 37, 387–399. American, November. Fthenakis, V.M., Kim, H.C., 2007. Greenhouse-gas emissions from solar electric- and Jacobson, M.Z., 2010. Short-term effects of controlling fossil-fuel soot, biofuel soot nuclear power: a lifecycle study. Energy Policy 35, 2549–2557. and gases, and methane on climate, Arctic ice, and air pollution health. Journal of Gaines, L., Nelson, P., 2009. Lithium-ion batteries: possible materials issues. In: Geophysical Research 115, D14209. doi:10.1029/2009JD013795. Paper Presented at The 13th International Battery Materials Recycling Seminar Janic, M., 2008. The potential of liquid hydrogen for the future ‘carbon-neutral’ air and Exhibit, Broward County Convention Center, Fort Lauderdale, Florida, March transport system. Transportation Research D 13, 428–435. 16–18, 2009. Jaskula, B.W., 2008. Lithium. 2007 Minerals Yearbook. U.S. Geological Survey, U.S. Gaines, L., Nelson, P., 2010. Lithium-ion batteries: examining material demand and Department of the Interior, August 2008. /http://minerals.usgs.gov/minerals/ recycling issues. Argonne National Laboratory, presented at 2010 TMS Annual pubs/commodity/lithium/myb1-2007-lithi.pdfS. Meeting and Exhibition, Seattle Washington, February 14–18, 2010. /http:// Joskow, P.L., Parsons, J.E., 2009. The economic future of nuclear power. Daedalus, www.transportation.anl.gov/pdfs/B/626.PDFS. Fall, 45–59.

Ginnebaugh, D.L., Liang, J., Jacobson, M.Z., 2010. Examining the temperature Kempton, W., Archer, C.L., Dhanju, A., Garvine, R.W., Jacobson, M.Z., 2007. Large CO2 dependence of ethanol (E85) versus gasoline emissions on air pollution with a reductions via matched to inherent storage in energy end- largely-explicit chemical mechanism. Atmospheric Environment 44, uses. Geophysical Research Letters 34, L02817. doi:10.1029/2006GL028016. 1192–1199, doi:10.1016/j.atmosenv.2009.12.024. Kessides, I.N., 2010. Nuclear power: understanding the economic risks and Glaser, A., Ramana, M.V., 2007. Weapon-grade plutonium production potential in uncertainties. Energy Policy 38, 3849–3864. the India prototype fast breeder reactor. Science and Global Security 15, 85–105. Koomey, J., Hultman, N.E., 2007. A reactor-level analysis of busbar costs for U.S. Gorman, S., 2009. As hybrid cars gobble rare metals, shortage looms. , August 31, nuclear plants, 1970–2005. Energy Policy 35, 5630–5642. 2009. /http://www.reuters.com/article/newsOne/idUSTRE57U02B20090831S (last Koroneos, C., Dompros, A., Roumbas, G., Moussiopoulos, N., 2005. Advantages of the accessed August 28, 2010). use of hydrogen fuel as compared to kerosone. Resources, Conservation, and Gordon, R.B., Bertram, M., Graedel, T.E., 2006. Metal stocks and sustainability. Recycling 44, 99–113. Proceedings of the National Academy of Science 103 (5), 1209–1214. Lefevre, M., Proietti, E., Jaouen, F., Dodelet, J.-P., 2009. Iron-based catalysts with Goto, H., Suzuki, Y., Nakamura, K., Watanabe, T., Guo, H.J., Ichinokura, O., 2005. A improved oxygen reduction activity in polymer electrolyte fuel cells. Science multipolar SR motor and its application in EV. Journal of Magnetism and 324, 71–74. Magnetic Materials 290–291, 1338–1342. Lenzen, M., 2008. Life cycle energy and greenhouse gas emissions of nuclear energy: Grubler, A., 2010. The costs of the French nuclear scale-up: a case of negative a review. Energy Conversion & Management 49, 2178–2199. learning by doing. Energy Policy 38, 5174–5188. Lifton, J., 2009. Braking wind: where’s the neodymium going to come from? Hageluken,¨ C., 2009. Umicore precious metals refining, Hanau, Germany, personal /http://www.glgroup.com/News/Braking-Wind–Wheres-the-Neodymium- communications via e-mail, October–November 2009). Going-To-Come-from–35041.htmlS. Hageluken,¨ C., Buchert, M., Ryan, P., 2009b. Materials flow of platinum group Lovins, A.B., Howe, B., 1992. Switched reluctance motors poised for rapid growth. metals in Germany. International Journal of Sustainable Manufacturing 1, Tech Update, TU-92-4, E-Source, a subsidiary of , 330–346. See also Hageluken,¨ C., Buchert, M., Ryan, P., 2006. Materials flow Boulder, Colorado, November 1992. of platinum group metals in Germany, Proceedings of LCE 2006, pp. 477–482. Lu, X., McElroy, M.B., Kiviluoma, J., 2009. Global potential for wind-generated /http://www.mech.kuleuven.be/lce2006/052.pdfS. electricity. Proceedings of the National Academy of Sciences 106, 10933. Hammond, G.P., 1996. Nuclear energy in the twenty-first century. Applied Energy doi:10.1073/pnas.0904101106. 54, 327–344. Macfarlane, A.M., Miller, M., 2007. Nuclear energy and uranium resources. Elements Hannum, W.H., Wade, D.C., McFarlane, H.F., Hill, R.N., 1997. Nonproliferation and 3, 185–192. safeguards aspects of the IFR. Progress in Nuclear Energy 31, 203–217. Maniaci, D.C., 2006. Operational performance prediction of a hydrogen-fueled Harding, J., 2007. Economics of Nuclear Power and Proliferation Risks in a Carbon- commercial transport. College of Engineering Research Symposium 2006, Constrained World. White paper, Nonproliferation Policy Education Center. Pennsylvania State University, University Park, Pennsylvania, March 2006. /http://www.npolicy.org/files/20070600-Harding-EconomicsNewNuclear /http://www.engr.psu.edu/symposium2006/papers/Session%203E%20-%20E Power.pdfS. nergy/Maniaci.docS. Hatch, G. /http://nucleargreen.blogspot.com/2009/01/jack-liftons-research-on-min Margonelli, Lisa, 2009. Clean energy’s dirty little secret The Atlantic, May 2009. eral.htmlS; /http://www.terramagnetica.com/2009/08/03/how-does-using-per /http://www.theatlantic.com/doc/200905/hybrid-cars-mineralsS. manent-magnets-make-wind-turbines-more-reliable/S; /http://www.terramag Maximum EV, 2009. Rare earths and neodymium, June 21, 2009. /http://max netica.com/2009/08/07/10-mw-and-beyond-are-superconductors-the-future-of- imumev.blogspot.com/S (last accessed August 28, 2010). wind-energy/S (2009). Meridian International Research, 2008. The Trouble with Lithium 2, Under the Horikawa, T., Miura, K., Itoh, M., Machida, K., 2006. Effective recycling for Nd–Fe–B Microscope, Martianville, France, May 29, 2008. /http://www.meridian-int-res. sintered magnet scraps. Journal of Alloys and Compounds 408–412, 1386–1390. com/Projects/Lithium_Microscope.pdfS. Horrigan, L., Lawrence, R.S., Walker, P., 2002. How can Miller, S.E., Sagan, S.D., 2009. Nuclear Power without Proliferation? Daedalus, Fall, address the environmental and human health harms of industrial agriculture. 7–18. Environmental Health Perspectives 110, 445–456. Mital, S.K., Gyekenyesi, J.Z., Arnold, S.M., Sullivan, R.M., Manderscheid, J.M., Murthy, Hedrick, J.B., 2009. Rare Earths, chapter 60 in 2007 Minerals Yearbook. U.S. P.L.N., 2006. Review of current state of the art and key design issues with Geological Survey, U.S. Department of Interior. /http://minerals.usgs.gov/ potential solutions for liquid hydrogen cryogenic storage tank structures for minerals/pubs/commodity/rare_earths/myb1-2007-raree.pdfS. aircraft applications. NASA/TM-2006-214346, National Aeronautics and Space M.Z. Jacobson, M.A. Delucchi / Energy Policy 39 (2011) 1154–1169 1169

Administration, , , , October 2006. /http:// Spiegel, R.J., 2004. Platinum and fuel cells. Transportation Research D 9, 357–371. gltrs.grc.nasa.gov/reports/2006/TM-2006-214346.pdfS. Santa Maria, M.R.V., Jacobson, M.Z., 2009. Investigating the effect of large wind farms Morcos, T., 2009. Harvesting wind power with (or without) permanent magnets. on energy in the atmosphere. Energies 2, 816–838. Magnetics and Business Technology, Summer, p. 26. /http://www.magneticsma Sun, Y., Delucchi, M., Ogden, J., July 2010. Platinum Price and Fuel-cell System Cost, gazine.com/images/PDFs/Online%20Issues/2009/Magnetics_Summer09.pdfS. manuscript. Institute of Transportation Studies, University of California, Davis. Mourogov, V., 2000. Role of nuclear energy for sustainable development. Progress in Takeda, O.T.H., Okabe, Umetsu, Y., 2006. Recovery of neodymium from a mixture of Nuclear Energy 37, 19–24. magnet scrap and other scrap. Journal of Alloys and Compounds 408-412, Mourogov, V., Fukuda, K., Kagramanian, V., 2002. The need for innovative nuclear 387–390. reactor and fuel cycle systems: strategy for development and future prospects. TIAX L.L.C., 2003. Platinum Availability and Economics for PEMFC Commercializa- Progress in Nuclear Energy 40, 285–299. tion. Report to the U.S. Department of Energy, DE-FC04-01AL67601, TIAX LLC, Niles, J.O., Brown, S., Pretty, J., Ball, A.S., Fay, J., 2002. Potential carbon mitigation and Cambridge, Massachusetts, December 2003. /http://www1.eere.energy.gov/ income in developing countries from changes in land use and management of hydrogenandfuelcells/pdfs/tiax_platinum.pdfS. agricultural and forest lands. Philosophical Transactions of the Royal Society of Tokimatsu, K., Fujino, J., Konishi, S., Ogawa, Y., Yamaji, K., 2003. Role of nuclear fusion London 360, 1621–1639. in future energy systems and the environment under future uncertainties. Nojoumi, H., Dincer, I., Naterer, G.F., 2009. greenhouse gas emissions assessment of Energy Policy 31, 775–797. hydrogen and kerosene-fueled aircraft propulsion. International Journal of Toyota, Hybrid synergy drive information terminal. /http://www.hybridsynergyd Hydrogen Energy 34, 1363–1369. rive.com/en/electric_motor.htmlS (last accessed August 28, 2010). Nuclear Power Daily,2009. Nuclear Safety Bodies Call for Redesign of EPR Reactor, Till, C.E., Chang, Y.I., Hannum, W.H., 1997. The integral fast reactor—an overview. November 2. /http://www.nuclearpowerdaily.com/reports/Nuclear_safety_bodies_ Progress in Nuclear Energy 31, 3–11. call_for_redesign_of_EPR_reactor_999.htmlS (last accessed August 28, 2010). Tilman, D., Hill, J., Lehman, C., 2006. Carbon-negative biofuels from low-input high- OICA, International Organization of Motor Vehicle Manufacturers, ‘‘2009 Production diversity grassland biomass. Science 314, 1598–1600. Statistics’’. /http://oica.net/category/production-statistics/S (accessed August Ullom, J., 1994. Enriched uranium versus plutonium: proliferant preferences in the 28, 2010. choice of fissile material. The Nonproliferation Review 1 (2), 1–15. Ongena, J., Van Oost, G., 2006. Energy for Future centuries: prospectsfor fusion power as United Kingdom Department of Transport (UKDOT), 2006. Platinum and hydrogen a future energy source. Transactions of Fusion Science and Technology 49, 3–15. for fuel cell vehicles, January 26, 2006. /http://www.dft.gov.uk/pgr/roads/ O’Rourke, F., Boyle, F., Reynolds, A., 2010. Tidal energy update 2009. Applied Energy environment/research/cqvcf/platinumandhydrogenforfuelce3838S. 87, 398–409. U.S. Department of Energy (USDOE), 2008a. Energy Efficiency and Renewable Pacala, S., Socolow, R., 2004. Stabilization wedges: solving the climate problem for Energy, 20% Wind Energy by 2030, Increasing Wind Energy’s Contribution to the next 50 years with current technologies. Science 305, 968–971. U.S. Electricity Supply. DOE/GO-102008-256, Washington, D.C. /http://20per Parsons-Brinckerhoff, Powering the Future: Mapping our Low-Carbon Path to 2050, centwind.org/20percent_wind_energy_report_revOct08.pdfS. December (2009). /http://www.pbpoweringthefuture.comS. U.S. Department of Energy (USDOE), 2008b. Concentrating Solar Power Commercial Penner, S.S., Seiser, R., Schultz, K.R., 2008. Steps toward passively safe, proliferation- Application Study: Reducing Water Consumption of Concentrating Solar Power resistant nuclear power. Progress in Energy and Combustion Science 34, , Report to Congress. /http://www1.eere.energy.gov/ 275–287. solar/pdfs/csp_water_study.pdfS (last accessed Jan. 24, 2010). Pickens, T.B., 2009. . /http://www.pickensplan.com/theplan/,%20Acces U.S. Geological Survey (USGS), 2009. Mineral Commodities Summaries 2009, U.S. sed%20Nov.%208,%202010S. Government Printing Office, Washington, D.C./http://minerals.usgs.gov/miner Piera, M., 2010. Sustainability issues in the development of nuclear fission energy. als/pubs/mcs/S. Energy Conversion and Management 51, 938–946. Von Hippel, F.N., 2008. Nuclear fuel recycling: more trouble than its worth. Scientific Price-Waterhouse-Coopers, 2010. 100% renewable electricity: a roadmap to 2050 American2008. /http://old.armscontrolcenter.org/policy/nonproliferation/arti for Europe and North Africa. /http://www.pwcwebcast.co.uk/dpliv_mu/ cles/nuclear_fuel_recycling/S. 100_percent_renewable_electricity.pdfS. Van Schaik, A., Reuter, M.A., 2004. The time-varying factors influencing the recycling Purushotham, D.S.C., Venugopal, V., Ramanujam, A., 2000. Nuclear fuel cycle: recent rate of products. Resources, Conservation, and Recycling 40, 301–328. developments and future directions. Journal of Radioanalytica and Nuclear Wadia, C., Alivisatos, A.P., Kammen, D.M., 2009. Materials availability expands the Chemistry 243, 199–203. opportunity for large-scale photovoltaics deployment. Environmental Science Ramana, M.V., 2009. Nuclear power: economic, safety, health, and environmental and Technology 43, 2072–2077. issues of near-term technologies. Annual Review of Environment and Resources Wall, E., Smit, B., 2005. Climate change adaption in light of sustainable agriculture. 34, 127–152. Journal of Sustainable Agriculture 27, 113–123. Reisman, L., 2009. China and rare earth metals: two sides to every story (Part one). Weisser, D., 2007. A guide to life-cycle greenhouse gas (GHG) emissions from /http://agmetalminer.com/2009/09/03/china-and-rare-earth-metals-–-two- electric supply technologies. Energy 32, 1543–1559. sides-to-every-story-part-one/S, September 2009. Westenberger, A., 2003. Hydrogen fuelled aircraft. In: Presented at the AIAA/ICAS Risen, J., 2010. U.S. identifies vast mineral riches in Afghanistan. New York Times, International Air and Space Symposium and Exposition: The Next 100 Years, June 13, 2010. /http://www.nytimes.com/2010/06/14/world/asia/14minerals. Dayton, Ohio, AIAA 2003-2880./http://www.aiaa.org/content.cfm?pageid=406 htmlS. &gTable=Paper&gID=3274S. Ritter, S.K., 2009. Future of metals. Chemical and Engineering News 87 (23), Wilburn, D.R., 2009. Material use in the United States—selected case studies for 53–57 /http://pubs.acs.org/cen/science/87/8723sci1.htmlS. cadmium, cobalt, lithium, and nickel in rechargeable batteries. Scientific Sovacool, B.K., 2008. Valuing the greenhouse gas emissions from nuclear power: a Investigations Report 2008-5141, U.S. Geological Survey, U.S. Department of critical survey. Energy Policy 36, 2940–2953. the Interior. /http://pubs.usgs.gov/sir/2008/5141/S. Sovacool, B.K., Watts, C., 2009. Going completely renewable: is it possible (let alone Wright, L., 2010. Lithium dreams: can Bolivia become the Saudi Arabia of the desirable)? The Electricity Journal 22 (4), 95–111. electric-car era? The New Yorker, March 22, 48–59. Sovacool, B.K., Sovacool, K.E., 2009. Identifying future electricity–water tradeoffs in Yang, C.-J., 2009. An impending platinum crisis and its implications for the future of the United States. Energy Policy 37, 2763–2773. the automobile. Energy Policy 37, 1805–1808.